Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,13 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+examples/ex3_concept.png filter=lfs diff=lfs merge=lfs -text
+examples/ex2_preview.png filter=lfs diff=lfs merge=lfs -text
+examples/ex1_preview.png filter=lfs diff=lfs merge=lfs -text
+examples/ex2_skin.png filter=lfs diff=lfs merge=lfs -text
+examples/ex3_skin.png filter=lfs diff=lfs merge=lfs -text
+examples/ex3_preview.png filter=lfs diff=lfs merge=lfs -text
+examples/ex2_concept.png filter=lfs diff=lfs merge=lfs -text
+examples/ex1_concept.png filter=lfs diff=lfs merge=lfs -text
+examples/ex1_skin.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,60 @@
+# BLOCKv0.5
+
+BLOCKv0.5 is an image-to-image model for converting a **3D Minecraft character preview** into a **2D skin texture file**.
+
+It is based on `black-forest-labs/FLUX.2-klein-base-9B` with merged LoRA weights and was trained with **[cond, target]** token order.  
+For best results, use the included custom pipeline: `Flux2KleinPipelineTrainOrder`.
+
+## What This Model Does
+
+- Input: a character preview image (RGB, square recommended)
+- Optional control: text prompt for style/details
+- Output: a generated skin texture image (PNG)
+
+## Quick Start
+
+```python
+import torch
+from PIL import Image
+from pipeline_flux2_klein_train_order import Flux2KleinPipelineTrainOrder
+
+model_id = "your-username/BLOCKv0.5"  # or local path
+
+pipe = Flux2KleinPipelineTrainOrder.from_pretrained(
+    model_id,
+    torch_dtype=torch.bfloat16,
+)
+pipe.to("cuda")
+
+preview = Image.open("examples/ex1_preview.png").convert("RGB").resize((1024, 1024))
+
+result = pipe(
+    prompt="clean minecraft skin texture, readable details, game-ready layout",
+    image=preview,
+    num_inference_steps=30,
+    guidance_scale=4.0,
+).images[0]
+
+result.save("generated_skin.png")
+```
+
+## Example Results
+
+| Preview | Generated Skin |
+|---|---|
+| ![ex1 preview](examples/ex1_preview.png) | ![ex1 skin](examples/ex1_skin.png) |
+| ![ex2 preview](examples/ex2_preview.png) | ![ex2 skin](examples/ex2_skin.png) |
+| ![ex3 preview](examples/ex3_preview.png) | ![ex3 skin](examples/ex3_skin.png) |
+
+## Notes
+
+- This checkpoint is designed for img2img skin generation, not general text-to-image generation.
+- The included custom pipeline matches the training token order and is recommended for inference.
+- The model is large; use a high-memory GPU for practical inference speed.
+
+## Dependencies
+
+- diffusers
+- transformers
+- torch
+- safetensors

model_index.json

@@ -0,0 +1,26 @@
+{
+  "_class_name": "Flux2KleinPipeline",
+  "_diffusers_version": "0.37.0.dev0",
+  "_name_or_path": "black-forest-labs/FLUX.2-klein-base-9B",
+  "is_distilled": false,
+  "scheduler": [
+    "diffusers",
+    "FlowMatchEulerDiscreteScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "Qwen3ForCausalLM"
+  ],
+  "tokenizer": [
+    "transformers",
+    "Qwen2Tokenizer"
+  ],
+  "transformer": [
+    "diffusers",
+    "Flux2Transformer2DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKLFlux2"
+  ]
+}

pipeline_flux2_klein_train_order.py

@@ -0,0 +1,163 @@
+# Copyright 2025 GenMC. Custom pipeline for [cond, target] token order (aligned with BLOCKv0.5 training).
+#
+# This pipeline overrides Flux2KleinPipeline's img2img behavior:
+#   - Default: [target, cond], output first part
+#   - Ours:    [cond, target], output last part
+#
+# Use: pipe = Flux2KleinPipelineTrainOrder.from_pretrained("path/to/BLOCKv0.5")
+#      image = pipe(prompt="...", image=cond_img).images[0]
+
+import contextlib
+import numpy as np
+import torch
+from PIL import Image
+
+from diffusers import Flux2KleinPipeline
+from diffusers.pipelines.flux2.pipeline_flux2_klein import compute_empirical_mu, retrieve_timesteps
+from diffusers.pipelines.flux2.pipeline_output import Flux2PipelineOutput
+
+
+def _maybe_cache_context(model, name: str):
+    fn = getattr(model, "cache_context", None)
+    if fn is None:
+        return contextlib.nullcontext()
+    return fn(name)
+
+
+class Flux2KleinPipelineTrainOrder(Flux2KleinPipeline):
+    """
+    Flux2KleinPipeline with [cond, target] token order for img2img (aligned with BLOCKv0.5 training).
+    When image is provided, uses train-order concat and output slicing.
+    """
+
+    @torch.no_grad()
+    def __call__(self, image=None, **kwargs):
+        if image is not None:
+            return self._call_img2img_train_order(image=image, **kwargs)
+        return super().__call__(image=image, **kwargs)
+
+    def _call_img2img_train_order(
+        self,
+        image,
+        prompt=None,
+        height=None,
+        width=None,
+        num_inference_steps=50,
+        guidance_scale=4.0,
+        generator=None,
+        **kwargs,
+    ):
+        """Img2img with [cond, target] order. Mirrors flux2_infer.generate_img2img_train_order."""
+        device = self._execution_device
+        tf_dtype = self.transformer.dtype
+        vae_dtype = self.vae.dtype
+
+        if not isinstance(image, list):
+            image = [image]
+        self.check_inputs(prompt=prompt, height=height, width=width, guidance_scale=guidance_scale)
+        self._guidance_scale = guidance_scale
+        self._attention_kwargs = kwargs.get("attention_kwargs")
+
+        batch_size = 1 if isinstance(prompt, str) else len(prompt)
+        prompt_embeds, text_ids = self.encode_prompt(prompt=prompt, device=device)
+        prompt_embeds = prompt_embeds.to(device=device, dtype=tf_dtype)
+        text_ids = text_ids.to(device=device)
+
+        do_cfg = (guidance_scale is not None) and (float(guidance_scale) > 1.0)
+        if do_cfg:
+            neg_embeds, neg_text_ids = self.encode_prompt(prompt="", device=device)
+            neg_embeds = neg_embeds.to(device=device, dtype=tf_dtype)
+            neg_text_ids = neg_text_ids.to(device=device)
+
+        # Preprocess cond image
+        img = image[0]
+        if hasattr(img, "size"):
+            h, w = height or img.size[1], width or img.size[0]
+        else:
+            h, w = height or 512, width or 512
+        cond_tensor = self.image_processor.preprocess(
+            img, height=h, width=w, resize_mode="crop"
+        ).to(device=device, dtype=vae_dtype)
+        image_latents, image_latent_ids = self.prepare_image_latents(
+            images=[cond_tensor], batch_size=batch_size, generator=generator, device=device, dtype=vae_dtype
+        )
+        image_latents = image_latents.to(device=device, dtype=tf_dtype)
+        image_latent_ids = image_latent_ids.clone()
+        image_latent_ids[..., 0] = 0  # cond t_index=0
+
+        # Init target latents
+        num_channels_latents = self.transformer.config.in_channels // 4
+        latents, latent_ids = self.prepare_latents(
+            batch_size=batch_size,
+            num_latents_channels=num_channels_latents,
+            height=h,
+            width=w,
+            dtype=vae_dtype,
+            device=device,
+            generator=generator,
+            latents=None,
+        )
+        latents = latents.to(device=device, dtype=tf_dtype)
+        latent_ids = latent_ids.clone()
+        latent_ids[..., 0] = 1  # target t_index=1
+
+        # Timesteps
+        sigmas = np.linspace(1.0, 1 / num_inference_steps, num_inference_steps)
+        if hasattr(self.scheduler.config, "use_flow_sigmas") and self.scheduler.config.use_flow_sigmas:
+            sigmas = None
+        mu = compute_empirical_mu(image_seq_len=latents.shape[1], num_steps=num_inference_steps)
+        timesteps, _ = retrieve_timesteps(self.scheduler, num_inference_steps, device, sigmas=sigmas, mu=mu)
+
+        if hasattr(self.scheduler, "set_begin_index"):
+            self.scheduler.set_begin_index(0)
+
+        attn_kwargs = getattr(self, "attention_kwargs", None) or getattr(self, "_attention_kwargs", None)
+
+        for t in timesteps:
+            timestep = t.expand(latents.shape[0]).to(latents.dtype)
+            latent_model_input = torch.cat([image_latents, latents], dim=1).to(tf_dtype)
+            latent_image_ids = torch.cat([image_latent_ids, latent_ids], dim=1)
+
+            with _maybe_cache_context(self.transformer, "cond"):
+                out_all = self.transformer(
+                    hidden_states=latent_model_input,
+                    timestep=timestep / 1000,
+                    guidance=None,
+                    encoder_hidden_states=prompt_embeds,
+                    txt_ids=text_ids,
+                    img_ids=latent_image_ids,
+                    joint_attention_kwargs=attn_kwargs,
+                    return_dict=False,
+                )[0]
+            noise_pred = out_all[:, -latents.size(1) :, :]
+
+            if do_cfg:
+                with _maybe_cache_context(self.transformer, "uncond"):
+                    neg_all = self.transformer(
+                        hidden_states=latent_model_input,
+                        timestep=timestep / 1000,
+                        guidance=None,
+                        encoder_hidden_states=neg_embeds,
+                        txt_ids=neg_text_ids,
+                        img_ids=latent_image_ids,
+                        joint_attention_kwargs=attn_kwargs,
+                        return_dict=False,
+                    )[0]
+                neg_pred = neg_all[:, -latents.size(1) :, :]
+                noise_pred = neg_pred + float(guidance_scale) * (noise_pred - neg_pred)
+
+            latents = self.scheduler.step(noise_pred, t, latents, return_dict=False)[0]
+
+        latents = self._unpack_latents_with_ids(latents, latent_ids)
+        latents_bn_mean = self.vae.bn.running_mean.view(1, -1, 1, 1).to(latents.device, latents.dtype)
+        latents_bn_std = torch.sqrt(
+            self.vae.bn.running_var.view(1, -1, 1, 1) + self.vae.config.batch_norm_eps
+        ).to(latents.device, latents.dtype)
+        latents = latents * latents_bn_std + latents_bn_mean
+        latents = self._unpatchify_latents(latents)
+        if latents.dtype != self.vae.dtype:
+            latents = latents.to(self.vae.dtype)
+        image_out = self.vae.decode(latents, return_dict=False)[0]
+        image_out = self.image_processor.postprocess(image_out, output_type="pil")
+        self.maybe_free_model_hooks()
+        return Flux2PipelineOutput(images=image_out)

scheduler/scheduler_config.json

@@ -0,0 +1,18 @@
+{
+  "_class_name": "FlowMatchEulerDiscreteScheduler",
+  "_diffusers_version": "0.37.0.dev0",
+  "base_image_seq_len": 256,
+  "base_shift": 0.5,
+  "invert_sigmas": false,
+  "max_image_seq_len": 4096,
+  "max_shift": 1.15,
+  "num_train_timesteps": 1000,
+  "shift": 3.0,
+  "shift_terminal": null,
+  "stochastic_sampling": false,
+  "time_shift_type": "exponential",
+  "use_beta_sigmas": false,
+  "use_dynamic_shifting": true,
+  "use_exponential_sigmas": false,
+  "use_karras_sigmas": false
+}

text_encoder/config.json

@@ -0,0 +1,71 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 12288,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 40960,
+  "max_window_layers": 36,
+  "model_type": "qwen3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "pad_token_id": null,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 1000000,
+    "rope_type": "default"
+  },
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "5.2.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

text_encoder/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "transformers_version": "5.2.0",
+  "use_cache": true
+}

text_encoder/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e620a91ee8ea327930fe1f64d26c16bb64445bede25a68a3b6f252df3490475
+size 16381517208

tokenizer/chat_template.jinja

@@ -0,0 +1,89 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
+size 11422650

tokenizer/tokenizer_config.json

@@ -0,0 +1,14 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "is_local": true,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

transformer/config.json

@@ -0,0 +1,24 @@
+{
+  "_class_name": "Flux2Transformer2DModel",
+  "_diffusers_version": "0.37.0.dev0",
+  "_name_or_path": "/home/guohq2021/.cache/huggingface/hub/models--black-forest-labs--FLUX.2-klein-base-9B/snapshots/32773329fbe7e81a90ef971740e8ba4b0364ecf3/transformer",
+  "attention_head_dim": 128,
+  "axes_dims_rope": [
+    32,
+    32,
+    32,
+    32
+  ],
+  "eps": 1e-06,
+  "guidance_embeds": false,
+  "in_channels": 128,
+  "joint_attention_dim": 12288,
+  "mlp_ratio": 3.0,
+  "num_attention_heads": 32,
+  "num_layers": 8,
+  "num_single_layers": 24,
+  "out_channels": null,
+  "patch_size": 1,
+  "rope_theta": 2000,
+  "timestep_guidance_channels": 256
+}

transformer/diffusion_pytorch_model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7f7afa7c84dbc9b145964de7a75be3aea9ad07fe6137c8af11016390f84d7c5
+size 9801069272

transformer/diffusion_pytorch_model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d891d8c65a14816e84148cb81d9455acd666f66ef703df497659ff0b1c1516ca
+size 8356121608

transformer/diffusion_pytorch_model.safetensors.index.json

@@ -0,0 +1,240 @@
+{
+  "metadata": {
+    "total_size": 18157162496
+  },
+  "weight_map": {
+    "context_embedder.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "double_stream_modulation_img.linear.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "double_stream_modulation_txt.linear.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "norm_out.linear.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "proj_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_stream_modulation.linear.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.0.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.0.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.0.attn.to_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.0.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.1.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.1.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.1.attn.to_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.1.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.10.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.10.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.10.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.10.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.11.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.11.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.11.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.11.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.12.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.12.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.12.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.12.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.13.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.13.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.13.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.13.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.14.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.14.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.14.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.14.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.15.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.15.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.15.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.15.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.16.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.16.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.16.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.16.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.17.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.17.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.17.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.17.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.18.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.18.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.18.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.18.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.19.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.19.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.19.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.19.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.2.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.2.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.2.attn.to_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.2.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.20.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.20.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.20.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.20.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.21.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.21.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.21.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.21.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.22.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.22.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.22.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.22.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.23.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.23.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.23.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.23.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.3.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.3.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.3.attn.to_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.3.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.4.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.4.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.4.attn.to_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.4.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "single_transformer_blocks.5.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.5.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.5.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.5.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.6.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.6.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.6.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.6.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.7.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.7.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.7.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.7.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.8.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.8.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.8.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.8.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.9.attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.9.attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.9.attn.to_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "single_transformer_blocks.9.attn.to_qkv_mlp_proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "time_guidance_embed.timestep_embedder.linear_1.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "time_guidance_embed.timestep_embedder.linear_2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff_context.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff_context.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff_context.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff_context.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff_context.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff_context.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff_context.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff_context.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff_context.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff_context.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff_context.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff_context.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff_context.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff_context.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff_context.linear_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff_context.linear_out.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "x_embedder.weight": "diffusion_pytorch_model-00001-of-00002.safetensors"
+  }
+}

vae/config.json

@@ -0,0 +1,40 @@
+{
+  "_class_name": "AutoencoderKLFlux2",
+  "_diffusers_version": "0.37.0.dev0",
+  "_name_or_path": "/home/guohq2021/.cache/huggingface/hub/models--black-forest-labs--FLUX.2-klein-base-9B/snapshots/32773329fbe7e81a90ef971740e8ba4b0364ecf3/vae",
+  "act_fn": "silu",
+  "batch_norm_eps": 0.0001,
+  "batch_norm_momentum": 0.1,
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 32,
+  "layers_per_block": 2,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "patch_size": [
+    2,
+    2
+  ],
+  "sample_size": 1024,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": true,
+  "use_quant_conv": true
+}

vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca70d2202afe6415bdbcb8793ba8cd99fd159cfe6192381504d6c4d3036e0f04
+size 168120878