Kijai

/

MiniMax-H3-experimental

like 234

~/ComfyUI w4a8 ⇡
comfyvenv ❯ python main.py --enable-manager --models-directory ~aero/comfy/ComfyUI/models/ --cache-none
[INFO] setup plugin alembic.autogenerate.schemas
[INFO] setup plugin alembic.autogenerate.tables
[INFO] setup plugin alembic.autogenerate.types
[INFO] setup plugin alembic.autogenerate.constraints
[INFO] setup plugin alembic.autogenerate.defaults
[INFO] setup plugin alembic.autogenerate.comments
[INFO] Setting models directory to: /home/aero/comfy/ComfyUI/models
[START] Security scan
[INFO] [ComfyUI-Manager] Using uv as Python module for pip operations.
Using Python 3.11.14 environment at: /home/aero/comfyvenv
[DONE] Security scan
** ComfyUI startup time: 2026-08-06 22:53:41.665
** Platform: Linux
** Python version: 3.11.14 (main, Nov 20 2025, 15:58:12) [GCC 15.2.1 20251112]
** Python executable: /home/aero/comfyvenv/bin/python
** ComfyUI Path: /home/aero/ComfyUI
** ComfyUI Base Folder Path: /home/aero/ComfyUI
** User directory: /home/aero/ComfyUI/user
** ComfyUI-Manager config path: /home/aero/ComfyUI/user/manager/config.ini
** Log path: /home/aero/ComfyUI/user/comfyui.log
Using Python 3.11.14 environment at: /home/aero/comfyvenv
Using Python 3.11.14 environment at: /home/aero/comfyvenv
[INFO] [PRE] ComfyUI-Manager
[INFO] Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1', 'apply_rope', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'int8_linear', 'quantize_and_rotate_rowwise', 'quantize_int8_rowwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'w4a8_int8_linear']}
[INFO] Found comfy_kitchen backend cuda: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'dequantize_w4a8_int8_weight', 'gemv_awq_w4a16', 'int8_linear', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'quantize_w4a8_int8_weight', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'rotate_int8_convrot_weight', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8', 'w4a8_int8_linear']}
[INFO] Found comfy_kitchen backend hip: {'available': False, 'disabled': False, 'unavailable_reason': 'PyTorch ROCm/HIP runtime not available', 'capabilities': []}
[INFO] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_embedding', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'dequantize_w4a8_int8_weight', 'gemv_awq_w4a16', 'int8_linear', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'quantize_w4a8_int8_weight', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'rotate_int8_convrot_weight', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8', 'w4a8_int8_linear']}
[INFO] Checkpoint files will always be loaded safely.
[INFO] Total VRAM 5800 MB, total RAM 15701 MB
[INFO] pytorch version: 2.11.0+cu130
[INFO] Set vram state to: NORMAL_VRAM
[INFO] Device: cuda:0 NVIDIA GeForce RTX 4050 Laptop GPU : cudaMallocAsync
[INFO] Using async weight offloading with 2 streams
[INFO] Enabled pinned memory 6280.0
[INFO] Using pytorch attention
[INFO] aimdo: /project/src/control.c:276:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 4050 Laptop GPU (VRAM: 5799 MB)
[INFO] DynamicVRAM support detected and enabled
[INFO] Python version: 3.11.14 (main, Nov 20 2025, 15:58:12) [GCC 15.2.1 20251112]
[INFO] ComfyUI version: 0.30.0
[INFO] comfy-aimdo version: 0.4.13
[INFO] comfy-kitchen version: 0.2.25
[INFO] comfyui-frontend-package version: 1.48.6
[INFO] comfyui-workflow-templates version: 0.11.31
[INFO] comfyui-embedded-docs version: 0.5.9
[INFO] comfy-aimdo version: 0.4.13
[WARNING] ________________________________________________________________________
WARNING WARNING WARNING WARNING WARNING

Installed comfy-kitchen version 0.2.25 is lower than the recommended version 0.2.26.

Please install the updated requirements.txt file by running:
/home/aero/comfyvenv/bin/python -m pip install -r /home/aero/ComfyUI/requirements.txt
If you are on the portable package you can run: update\update_comfyui.bat to solve this problem.

[INFO] [Prompt Server] web root: /home/aero/comfyvenv/lib/python3.11/site-packages/comfyui_frontend_package/static
[INFO] Asset seeder disabled
[INFO] [START] ComfyUI-Manager
[ComfyUI-Manager] Using GitPython backend
[INFO] [ComfyUI-Manager] network_mode: public
[WARNING] [ComfyUI-Manager] The matrix sharing feature has been disabled because the matrix-nio dependency is not installed.
To use this feature, please run the following command:
/home/aero/comfyvenv/bin/python -m pip install matrix-nio

[INFO] No OpenGL_accelerate module loaded: No module named 'OpenGL_accelerate'
[MiniMaxH3-Cache] Applied patch to "comfy.ldm.minimax.model.MiniMaxH3Model"
[INFO]
Import times for custom nodes:
[INFO] 0.0 seconds: /home/aero/ComfyUI/custom_nodes/websocket_image_save.py
[INFO] 0.0 seconds: /home/aero/ComfyUI/custom_nodes/ComfyUI-MiniMax-H3-Turbo
[INFO] 0.0 seconds: /home/aero/ComfyUI/custom_nodes/ComfyUI-MiniMaxH3-Cache
[INFO] 0.0 seconds: /home/aero/ComfyUI/custom_nodes/cellocut_native
[INFO] 0.0 seconds: /home/aero/ComfyUI/custom_nodes/ComfyUI-SolAttn_triton
[INFO] 0.1 seconds: /home/aero/ComfyUI/custom_nodes/ComfyUI-KJNodes
[INFO]
[INFO] Context impl SQLiteImpl.
[INFO] Will assume non-transactional DDL.
[INFO] Disabling intermediate node cache.
[WARNING] ________________________________________________________________________
WARNING WARNING WARNING WARNING WARNING

Installed comfy-kitchen version 0.2.25 is lower than the recommended version 0.2.26.

Please install the updated requirements.txt file by running:
/home/aero/comfyvenv/bin/python -m pip install -r /home/aero/ComfyUI/requirements.txt
If you are on the portable package you can run: update\update_comfyui.bat to solve this problem.

[INFO] Starting server

[INFO] To see the GUI go to: http://127.0.0.1:8188
[INFO] [ComfyUI-Manager] The ComfyRegistry cache update is still in progress, so an outdated cache is being used.
FETCH DATA from: /home/aero/comfyvenv/lib/python3.11/site-packages/comfyui_manager/custom-node-list.json [DONE]
[INFO] got prompt
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float32
[INFO] Found quantization metadata version 1
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float16
[INFO] Found quantization metadata version 1
[INFO] Using MixedPrecisionOps for text encoder
[INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
[INFO] Requested to load MiniMaxH3TEModel_
[INFO] Model MiniMaxH3TEModel_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB.
[INFO] Found quantization metadata version 1
[INFO] Detected mixed precision quantization
[INFO] Using mixed precision operations
[INFO] Native ops: asym_w4a8_int8, float8_e5m2, convrot_w4a4, float8_e4m3fn, int8_tensorwise , emulated ops: nvfp4, mxfp8
[INFO] model weight dtype torch.bfloat16, manual cast: torch.bfloat16
[INFO] model_type FLOW
[INFO] Using sage attention mode: auto
[INFO] Applying MiniMax H3 Memory Efficient Sage Attention Patch to all transformer blocks
[INFO] [sol_attn] chaining onto an existing attention override; Sol-Attn takes first refusal and delegates everything else to it
[INFO] [sol_attn] composed with 50 object-patched attention forward(s) (e.g. diffusion_model.blocks.0.attn.forward): Sol-Attn takes eligible self-attention calls, the existing patch keeps the rest
[INFO] Requested to load MiniMaxH3
[INFO] 0 models unloaded.
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 11956MB Staged. 208 patches attached. Force pre-loaded 210 weights: 1175 KB.
[MiniMaxH3-Cache] Skipped 0/1 steps (1.00x speedup).
[ERROR] !!! Exception during processing !!! Allocation on device 0 would exceed allowed memory. (out of memory)
Currently allocated : 4.22 GiB
Requested : 441.00 MiB
Device limit : 5.66 GiB
Free (according to CUDA): 10.12 MiB
PyTorch limit (set by user-supplied memory fraction)
: 17179869184.00 GiB
[ERROR] Traceback (most recent call last):
File "/home/aero/ComfyUI/execution.py", line 545, in execute
output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/execution.py", line 344, in get_output_data
return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/execution.py", line 318, in _async_map_node_over_list
await process_inputs(input_dict, i)
File "/home/aero/ComfyUI/execution.py", line 306, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy_api/internal/init.py", line 149, in wrapped_func
return method(locked_class, **inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy_api/latest/_io.py", line 1935, in EXECUTE_NORMALIZED
to_return = cls.execute(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy_extras/nodes_custom_sampler.py", line 1049, in execute
samples = guider.sample(noise.generate_noise(latent), latent_image, sampler, sigmas, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise.seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 1333, in sample
output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/patcher_extension.py", line 114, in execute
return self.wrappers[self.idx](self, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-MiniMaxH3-Cache/nodes.py", line 282, in call
return sample_fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/patcher_extension.py", line 106, in call
return new_executor.execute(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 1260, in outer_sample
output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 1235, in inner_sample
samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 1005, in sample
samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/utils/_contextlib.py", line 124, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-MiniMax-H3-Turbo/init.py", line 74, in _turbo_sampler
denoised = model(x, sigmas[i] * s_in, **extra_args)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 640, in call
out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 1208, in call
return self.outer_predict_noise(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 1215, in outer_predict_noise
).execute(x, timestep, model_options, seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 1218, in predict_noise
return sampling_function(self.inner_model, x, timestep, self.conds.get("negative", None), self.conds.get("positive", None), self.cfg, model_options=model_options, seed=seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 620, in sampling_function
out = calc_cond_batch(model, conds, x, timestep, model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 211, in calc_cond_batch
return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 219, in _calc_cond_batch_outer
return executor.execute(model, conds, x_in, timestep, model_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/samplers.py", line 335, in calc_cond_batch
output = model.apply_model(input_x, timestep, **c).chunk(batch_chunks)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/model_base.py", line 196, in apply_model
return comfy.patcher_extension.WrapperExecutor.new_class_executor(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/model_base.py", line 240, in _apply_model
model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/ldm/minimax/model.py", line 499, in forward
return comfy.patcher_extension.WrapperExecutor.new_class_executor(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/patcher_extension.py", line 113, in execute
return self.original(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-SolAttn_triton/_morton_h3.py", line 129, in _forward
return original_forward(x, timestep, context,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-MiniMaxH3-Cache/init.py", line 160, in patched_forward
h = blocks_replace[("block_loop", 0)](
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-MiniMaxH3-Cache/nodes.py", line 214, in __call__
res = original_block(args)
^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-MiniMaxH3-Cache/init.py", line 158, in block_loop_wrap
return {"img": self._run_blocks(args["img"], args["t_emb"], args["mod_segments"], args["rope_freqs"],
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-MiniMaxH3-Cache/init.py", line 44, in _run_blocks
h = block(h, t_emb, mod_segments, rope_freqs, transformer_options=transformer_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1885, in _call_impl
return inner()
^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1833, in inner
result = forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/ldm/minimax/model.py", line 270, in forward
x = _mod_gate(x, gate_msa, self.attn(h, rope_freqs=rope_freqs, transformer_options=transformer_options), mod_segments)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-SolAttn_triton/init.py", line 344, in forward
return patched_forward(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/custom_nodes/ComfyUI-KJNodes/nodes/ltxv_nodes.py", line 2078, in minimax_sageattn_forward
q, k, v = self.qkv_proj(x).split(self.heads * self.head_dim, dim=-1)
^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1779, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1790, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/ops.py", line 1425, in forward
output = self.forward_comfy_cast_weights(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/ops.py", line 1343, in forward_comfy_cast_weights
weight, bias, offload_stream = cast_bias_weight(
^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/ops.py", line 374, in cast_bias_weight
weight, bias = resolve_cast_module_with_vbar(s, dtype, device, bias_dtype, compute_dtype, want_requant)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/ops.py", line 321, in resolve_cast_module_with_vbar
weight = post_cast(s, "weight", weight, dtype, prefetch["resident"], update_weight)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/ComfyUI/comfy/ops.py", line 309, in post_cast
y = orig.requantize_from_float(x, scale="recalculate", stochastic_rounding=seed)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/comfy_kitchen/tensor/base.py", line 309, in requantize_from_float
return type(self).from_float(tensor, self._layout_cls, **qkwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/comfy_kitchen/tensor/base.py", line 239, in from_float
qdata, params = get_layout_class(layout_cls).quantize(tensor, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/comfy_kitchen/tensor/w4a8_int8.py", line 183, in quantize
qdata, s_rel, s_channel, correction, codebook_tensor = quantize_w4a8_int8_weight(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/comfy_kitchen/tensor/w4a8_int8.py", line 55, in quantize_w4a8_int8_weight
return impl(**kwargs)
^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/comfy_kitchen/backends/cuda/init.py", line 1371, in quantize_w4a8_int8_weight
return _quantize_rotated_w4a8_int8_weight(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/comfy_kitchen/backends/eager/w4a8_int8.py", line 125, in _quantize_rotated_w4a8_int8_weight
quantized = _assign_codes(normalized, codebook_tensor)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/aero/comfyvenv/lib/python3.11/site-packages/comfy_kitchen/backends/eager/w4a8_int8.py", line 17, in _assign_codes
distance = (normalized - codebook[index]).abs()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
torch.OutOfMemoryError: Allocation on device 0 would exceed allowed memory. (out of memory)
Currently allocated : 4.22 GiB
Requested : 441.00 MiB
Device limit : 5.66 GiB
Free (according to CUDA): 10.12 MiB
PyTorch limit (set by user-supplied memory fraction)
: 17179869184.00 GiB

[INFO] Memory summary:
|===========================================================================|

[ERROR] Got an OOM, unloading all loaded models.
[INFO] Prompt executed in 29.12 seconds