diff --git a/3drawers/gen_flex/chunk_001/_logs/b1_ep00000000.log b/3drawers/gen_flex/chunk_001/_logs/b1_ep00000000.log index 4ee0033da778db0ddc51e30c893c3bb78869d0e7..8b821d7e26327725c66bf5c1970c8a0a38411386 100644 --- a/3drawers/gen_flex/chunk_001/_logs/b1_ep00000000.log +++ b/3drawers/gen_flex/chunk_001/_logs/b1_ep00000000.log @@ -1,9 +1,3 @@ -/svl/u/ravenh/miniconda3/envs/mlspaces/lib/python3.11/site-packages/torch/cuda/__init__.py:287: UserWarning: -NVIDIA RTX PRO 6000 Blackwell Server Edition with CUDA capability sm_120 is not compatible with the current PyTorch installation. -The current PyTorch install supports CUDA capabilities sm_50 sm_60 sm_70 sm_75 sm_80 sm_86 sm_90. -If you want to use the NVIDIA RTX PRO 6000 Blackwell Server Edition GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/ - - warnings.warn( Views: ['ext1', 'wrist'] Prompt: This video shows in the background, a residential mudroom with an entryway area, with soft warm gray walls and white wainscot panels, light yellow pine plank floor slightly worn, and a row of black iron coat hooks holding a denim jacket on the wall. In the foreground, there is a warm taupe-painted three-drawer dresser with matte gold bar pulls, and a pair of leather boots and a folded scarf rest on a low wooden bench beside it. The robot is instructed to open the top drawer of the dresser. TeaCache thresh: 0.1 @@ -40,34 +34,76 @@ Loaded model: { } Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth" Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. -Traceback (most recent call last): - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 154, in - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=274 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=276 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=276 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=267 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=266 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=266 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=277 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=301 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=260 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=249 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=274 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=279 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=277 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=261 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=262 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=281 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=264 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=276 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=257 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=285 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=264 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=292 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=278 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=266 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=275 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=261 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=263 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=258 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=301 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=279 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=301 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=294 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=278 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=263 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=264 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=301 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=268 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=269 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=276 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=279 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=253 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=262 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=265 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=301 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=301 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=301 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=256 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=268 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=275 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=271 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=267 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=301 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=265 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=273 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=260 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=271 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=277 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=269 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00 - main() - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/sim2realgen/inference_long_video_fast.py", line 90, in main - pipe = WanVideoPipeline.from_pretrained( - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/pipelines/wan_video.py", line 134, in from_pretrained - model_pool = pipe.download_and_load_models(model_configs, vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/diffusion/base_pipeline.py", line 293, in download_and_load_models - model_pool.auto_load_model( - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 70, in auto_load_model - model = self.load_model_file(config, path, vram_config, vram_limit=vram_limit) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/models/model_loader.py", line 40, in load_model_file - model = load_model( - ^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/loader/model.py", line 46, in load_model - state_dict = state_dict_converter(state_dict) - ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/utils/state_dict_converters/wan_video_vae.py", line 6, in WanVideoVAEStateDictConverter - state_dict_['model.' + name] = state_dict[name] - ~~~~~~~~~~^^^^^^ - File "/svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/diffsynth/core/vram/disk_map.py", line 64, in __getitem__ - param = param.to(self.torch_dtype) - ^^^^^^^^^^^^^^^^^^^^^^^^^^ -RuntimeError: CUDA error: no kernel image is available for execution on the device -CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. -For debugging consider passing CUDA_LAUNCH_BLOCKING=1 -Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. +Loaded model: { + "model_name": "wan_video_vae", + "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE", + "extra_kwargs": null +} +Loading models from: "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth" +Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file. +Loaded model: { + "model_name": "wan_video_image_encoder", + "model_class": "diffsynth.models.wan_video_image_encoder.WanImageEncoder", + "extra_kwargs": null +} +Using wan_video_text_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_t5_umt5-xxl-enc-bf16.pth". +Using wan_video_dit from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/diffusion_pytorch_model.safetensors". +Using wan_video_vae from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/Wan2.1_VAE.pth". +Using wan_video_image_encoder from "./models/PAI/Wan2.1-Fun-V1.1-1.3B-Control/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth". +No wan_video_motion_controller models available. This is not an error. +No wan_video_vace models available. This is not an error. +No wan_video_vap models available. This is not an error. +No wans2v_audio_encoder models available. This is not an error. +No wan_video_animate_adapter models available. This is not an error. +Loading LoRA from: /svl/u/ravenh/renderscale/Sim2RealGen_Video_Model/lora_ckpts/Wan2.1-Fun-V1.1-1.3B-Mixed-DROID-AGIBOT-3View-DualControl-PrevFrame-droidnew_continue_lora/step-5294.safetensors +300 tensors are fused by LoRA. Fused LoRA layers cannot be cleared by `pipe.clear_lora()`. +Setting up multi-view pipeline... + +===== source: video0 ===== + total_frames=264 num_clips=4 + VAE encoding: 0%| | 0/1 [00:00