Upload converted Nunchaku Lite pipeline

.gitattributes

@@ -33,3 +33,5 @@ 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
+output_comparison.png filter=lfs diff=lfs merge=lfs -text
+tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,60 @@
+---
+license: apache-2.0
+base_model: Tongyi-MAI/Z-Image-Turbo
+pipeline_tag: text-to-image
+library_name: diffusers
+tags:
+  - z-image
+  - nunchaku
+  - svdquant
+  - int4
+  - quantization
+---
+
+# Z-Image-Turbo Nunchaku Lite INT4 r32
+
+Diffusers-loadable conversion of:
+
+- Base model: `Tongyi-MAI/Z-Image-Turbo`
+- Source repo: `nunchaku-ai/nunchaku-z-image-turbo`
+- Source checkpoint: `svdq-int4_r32-z-image-turbo.safetensors`
+
+The transformer uses `quant_method: nunchaku_lite`, INT4 SVDQ, group size 64, rank 32, and 238 quantized targets. Packed fused QKV and SwiGLU projections are split in logical layout into the stock Z-Image graph. The Qwen3 text encoder is packaged with BitsAndBytes 4-bit NF4.
+
+## Benchmark
+
+| Checkpoint | Latency | Max VRAM |
+| --- | ---: | ---: |
+| Converted Diffusers Nunchaku Lite INT4 r32 + BNB4 Qwen3 | 6.03 s (stdev 0.00 s) | 12.01 GiB |
+
+RTX 5090, 1024×1024, 9 scheduler steps (8 DiT forwards), guidance scale 0, seed 42, one warmup and three measured runs. VRAM is peak total device usage sampled through `nvidia-smi`. The native row uses the base model's BF16 Qwen3 encoder.
+
+## Output Comparison
+
+![Native reference (left) and converted output (right)](output_comparison.png)
+
+Both images use the same prompt, seed, scheduler, resolution, and step count. Native Nunchaku 1.x selects NVFP4 kernels on Blackwell, so the comparison uses native NVFP4 as a visual reference only (cross-precision MAE 23.37, RMSE 30.81).
+
+## Run
+
+Requires the Hugging Face `kernels` package and an NVIDIA Turing, Ampere, Ada, or Blackwell (not Hopper) GPU.
+
+```python
+import torch
+from diffusers import ZImagePipeline
+
+pipe = ZImagePipeline.from_pretrained(
+    "lite-infer/z-image-turbo-nunchaku-lite-int4_r32-bnb4-text-encoder",
+    torch_dtype=torch.bfloat16,
+).to("cuda")
+
+image = pipe(
+    prompt='A cinematic portrait of a red fox in a misty forest at sunrise, detailed fur, volumetric light',
+    height=1024,
+    width=1024,
+    num_inference_steps=9,
+    guidance_scale=0.0,
+    generator=torch.Generator("cuda").manual_seed(42),
+).images[0]
+image.save("output.png")
+```

model_index.json

@@ -0,0 +1,24 @@
+{
+    "_class_name": "ZImagePipeline",
+    "_diffusers_version": "0.36.0.dev0",
+    "scheduler": [
+        "diffusers",
+        "FlowMatchEulerDiscreteScheduler"
+    ],
+    "text_encoder": [
+        "transformers",
+        "Qwen3Model"
+    ],
+    "tokenizer": [
+        "transformers",
+        "Qwen2Tokenizer"
+    ],
+    "transformer": [
+        "diffusers",
+        "ZImageTransformer2DModel"
+    ],
+    "vae": [
+        "diffusers",
+        "AutoencoderKL"
+    ]
+}

scheduler/scheduler_config.json

@@ -0,0 +1,7 @@
+{
+  "_class_name": "FlowMatchEulerDiscreteScheduler",
+  "_diffusers_version": "0.36.0.dev0",
+  "num_train_timesteps": 1000,
+  "use_dynamic_shifting": false,
+  "shift": 3.0
+}

text_encoder/config.json

@@ -0,0 +1,86 @@
+{
+  "architectures": [
+    "Qwen3Model"
+  ],
+  "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": 2560,
+  "initializer_range": 0.02,
+  "intermediate_size": 9728,
+  "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,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "nf4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 1000000,
+    "rope_type": "default"
+  },
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.13.1",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

text_encoder/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46738664bad81b157a67ed7c46f20c233172a821cc6694f6518392c16becf479
+size 2822212864

tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

tokenizer/tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# 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>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\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\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- 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>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if message.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

transformer/config.json

@@ -0,0 +1,280 @@
+{
+  "_class_name": "ZImageTransformer2DModel",
+  "_diffusers_version": "0.36.0.dev0",
+  "all_f_patch_size": [
+    1
+  ],
+  "all_patch_size": [
+    2
+  ],
+  "axes_dims": [
+    32,
+    48,
+    48
+  ],
+  "axes_lens": [
+    1536,
+    512,
+    512
+  ],
+  "cap_feat_dim": 2560,
+  "dim": 3840,
+  "in_channels": 16,
+  "n_heads": 30,
+  "n_kv_heads": 30,
+  "n_layers": 30,
+  "n_refiner_layers": 2,
+  "norm_eps": 1e-05,
+  "qk_norm": true,
+  "rope_theta": 256.0,
+  "t_scale": 1000.0,
+  "quantization_config": {
+    "quant_method": "nunchaku_lite",
+    "compute_dtype": "bfloat16",
+    "svdq_w4a4": {
+      "precision": "int4",
+      "group_size": 64,
+      "rank": 32,
+      "targets": [
+        "context_refiner.0.attention.to_k",
+        "context_refiner.0.attention.to_out.0",
+        "context_refiner.0.attention.to_q",
+        "context_refiner.0.attention.to_v",
+        "context_refiner.0.feed_forward.w1",
+        "context_refiner.0.feed_forward.w2",
+        "context_refiner.0.feed_forward.w3",
+        "context_refiner.1.attention.to_k",
+        "context_refiner.1.attention.to_out.0",
+        "context_refiner.1.attention.to_q",
+        "context_refiner.1.attention.to_v",
+        "context_refiner.1.feed_forward.w1",
+        "context_refiner.1.feed_forward.w2",
+        "context_refiner.1.feed_forward.w3",
+        "layers.0.attention.to_k",
+        "layers.0.attention.to_out.0",
+        "layers.0.attention.to_q",
+        "layers.0.attention.to_v",
+        "layers.0.feed_forward.w1",
+        "layers.0.feed_forward.w2",
+        "layers.0.feed_forward.w3",
+        "layers.1.attention.to_k",
+        "layers.1.attention.to_out.0",
+        "layers.1.attention.to_q",
+        "layers.1.attention.to_v",
+        "layers.1.feed_forward.w1",
+        "layers.1.feed_forward.w2",
+        "layers.1.feed_forward.w3",
+        "layers.10.attention.to_k",
+        "layers.10.attention.to_out.0",
+        "layers.10.attention.to_q",
+        "layers.10.attention.to_v",
+        "layers.10.feed_forward.w1",
+        "layers.10.feed_forward.w2",
+        "layers.10.feed_forward.w3",
+        "layers.11.attention.to_k",
+        "layers.11.attention.to_out.0",
+        "layers.11.attention.to_q",
+        "layers.11.attention.to_v",
+        "layers.11.feed_forward.w1",
+        "layers.11.feed_forward.w2",
+        "layers.11.feed_forward.w3",
+        "layers.12.attention.to_k",
+        "layers.12.attention.to_out.0",
+        "layers.12.attention.to_q",
+        "layers.12.attention.to_v",
+        "layers.12.feed_forward.w1",
+        "layers.12.feed_forward.w2",
+        "layers.12.feed_forward.w3",
+        "layers.13.attention.to_k",
+        "layers.13.attention.to_out.0",
+        "layers.13.attention.to_q",
+        "layers.13.attention.to_v",
+        "layers.13.feed_forward.w1",
+        "layers.13.feed_forward.w2",
+        "layers.13.feed_forward.w3",
+        "layers.14.attention.to_k",
+        "layers.14.attention.to_out.0",
+        "layers.14.attention.to_q",
+        "layers.14.attention.to_v",
+        "layers.14.feed_forward.w1",
+        "layers.14.feed_forward.w2",
+        "layers.14.feed_forward.w3",
+        "layers.15.attention.to_k",
+        "layers.15.attention.to_out.0",
+        "layers.15.attention.to_q",
+        "layers.15.attention.to_v",
+        "layers.15.feed_forward.w1",
+        "layers.15.feed_forward.w2",
+        "layers.15.feed_forward.w3",
+        "layers.16.attention.to_k",
+        "layers.16.attention.to_out.0",
+        "layers.16.attention.to_q",
+        "layers.16.attention.to_v",
+        "layers.16.feed_forward.w1",
+        "layers.16.feed_forward.w2",
+        "layers.16.feed_forward.w3",
+        "layers.17.attention.to_k",
+        "layers.17.attention.to_out.0",
+        "layers.17.attention.to_q",
+        "layers.17.attention.to_v",
+        "layers.17.feed_forward.w1",
+        "layers.17.feed_forward.w2",
+        "layers.17.feed_forward.w3",
+        "layers.18.attention.to_k",
+        "layers.18.attention.to_out.0",
+        "layers.18.attention.to_q",
+        "layers.18.attention.to_v",
+        "layers.18.feed_forward.w1",
+        "layers.18.feed_forward.w2",
+        "layers.18.feed_forward.w3",
+        "layers.19.attention.to_k",
+        "layers.19.attention.to_out.0",
+        "layers.19.attention.to_q",
+        "layers.19.attention.to_v",
+        "layers.19.feed_forward.w1",
+        "layers.19.feed_forward.w2",
+        "layers.19.feed_forward.w3",
+        "layers.2.attention.to_k",
+        "layers.2.attention.to_out.0",
+        "layers.2.attention.to_q",
+        "layers.2.attention.to_v",
+        "layers.2.feed_forward.w1",
+        "layers.2.feed_forward.w2",
+        "layers.2.feed_forward.w3",
+        "layers.20.attention.to_k",
+        "layers.20.attention.to_out.0",
+        "layers.20.attention.to_q",
+        "layers.20.attention.to_v",
+        "layers.20.feed_forward.w1",
+        "layers.20.feed_forward.w2",
+        "layers.20.feed_forward.w3",
+        "layers.21.attention.to_k",
+        "layers.21.attention.to_out.0",
+        "layers.21.attention.to_q",
+        "layers.21.attention.to_v",
+        "layers.21.feed_forward.w1",
+        "layers.21.feed_forward.w2",
+        "layers.21.feed_forward.w3",
+        "layers.22.attention.to_k",
+        "layers.22.attention.to_out.0",
+        "layers.22.attention.to_q",
+        "layers.22.attention.to_v",
+        "layers.22.feed_forward.w1",
+        "layers.22.feed_forward.w2",
+        "layers.22.feed_forward.w3",
+        "layers.23.attention.to_k",
+        "layers.23.attention.to_out.0",
+        "layers.23.attention.to_q",
+        "layers.23.attention.to_v",
+        "layers.23.feed_forward.w1",
+        "layers.23.feed_forward.w2",
+        "layers.23.feed_forward.w3",
+        "layers.24.attention.to_k",
+        "layers.24.attention.to_out.0",
+        "layers.24.attention.to_q",
+        "layers.24.attention.to_v",
+        "layers.24.feed_forward.w1",
+        "layers.24.feed_forward.w2",
+        "layers.24.feed_forward.w3",
+        "layers.25.attention.to_k",
+        "layers.25.attention.to_out.0",
+        "layers.25.attention.to_q",
+        "layers.25.attention.to_v",
+        "layers.25.feed_forward.w1",
+        "layers.25.feed_forward.w2",
+        "layers.25.feed_forward.w3",
+        "layers.26.attention.to_k",
+        "layers.26.attention.to_out.0",
+        "layers.26.attention.to_q",
+        "layers.26.attention.to_v",
+        "layers.26.feed_forward.w1",
+        "layers.26.feed_forward.w2",
+        "layers.26.feed_forward.w3",
+        "layers.27.attention.to_k",
+        "layers.27.attention.to_out.0",
+        "layers.27.attention.to_q",
+        "layers.27.attention.to_v",
+        "layers.27.feed_forward.w1",
+        "layers.27.feed_forward.w2",
+        "layers.27.feed_forward.w3",
+        "layers.28.attention.to_k",
+        "layers.28.attention.to_out.0",
+        "layers.28.attention.to_q",
+        "layers.28.attention.to_v",
+        "layers.28.feed_forward.w1",
+        "layers.28.feed_forward.w2",
+        "layers.28.feed_forward.w3",
+        "layers.29.attention.to_k",
+        "layers.29.attention.to_out.0",
+        "layers.29.attention.to_q",
+        "layers.29.attention.to_v",
+        "layers.29.feed_forward.w1",
+        "layers.29.feed_forward.w2",
+        "layers.29.feed_forward.w3",
+        "layers.3.attention.to_k",
+        "layers.3.attention.to_out.0",
+        "layers.3.attention.to_q",
+        "layers.3.attention.to_v",
+        "layers.3.feed_forward.w1",
+        "layers.3.feed_forward.w2",
+        "layers.3.feed_forward.w3",
+        "layers.4.attention.to_k",
+        "layers.4.attention.to_out.0",
+        "layers.4.attention.to_q",
+        "layers.4.attention.to_v",
+        "layers.4.feed_forward.w1",
+        "layers.4.feed_forward.w2",
+        "layers.4.feed_forward.w3",
+        "layers.5.attention.to_k",
+        "layers.5.attention.to_out.0",
+        "layers.5.attention.to_q",
+        "layers.5.attention.to_v",
+        "layers.5.feed_forward.w1",
+        "layers.5.feed_forward.w2",
+        "layers.5.feed_forward.w3",
+        "layers.6.attention.to_k",
+        "layers.6.attention.to_out.0",
+        "layers.6.attention.to_q",
+        "layers.6.attention.to_v",
+        "layers.6.feed_forward.w1",
+        "layers.6.feed_forward.w2",
+        "layers.6.feed_forward.w3",
+        "layers.7.attention.to_k",
+        "layers.7.attention.to_out.0",
+        "layers.7.attention.to_q",
+        "layers.7.attention.to_v",
+        "layers.7.feed_forward.w1",
+        "layers.7.feed_forward.w2",
+        "layers.7.feed_forward.w3",
+        "layers.8.attention.to_k",
+        "layers.8.attention.to_out.0",
+        "layers.8.attention.to_q",
+        "layers.8.attention.to_v",
+        "layers.8.feed_forward.w1",
+        "layers.8.feed_forward.w2",
+        "layers.8.feed_forward.w3",
+        "layers.9.attention.to_k",
+        "layers.9.attention.to_out.0",
+        "layers.9.attention.to_q",
+        "layers.9.attention.to_v",
+        "layers.9.feed_forward.w1",
+        "layers.9.feed_forward.w2",
+        "layers.9.feed_forward.w3",
+        "noise_refiner.0.attention.to_k",
+        "noise_refiner.0.attention.to_out.0",
+        "noise_refiner.0.attention.to_q",
+        "noise_refiner.0.attention.to_v",
+        "noise_refiner.0.feed_forward.w1",
+        "noise_refiner.0.feed_forward.w2",
+        "noise_refiner.0.feed_forward.w3",
+        "noise_refiner.1.attention.to_k",
+        "noise_refiner.1.attention.to_out.0",
+        "noise_refiner.1.attention.to_q",
+        "noise_refiner.1.attention.to_v",
+        "noise_refiner.1.feed_forward.w1",
+        "noise_refiner.1.feed_forward.w2",
+        "noise_refiner.1.feed_forward.w3"
+      ]
+    }
+  }
+}

transformer/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:093a286e6d8849ba15808221c37b43b16e6b5f5b40ecc0b6661e3027c488ec49
+size 3634700712

vae/config.json

@@ -0,0 +1,38 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.36.0.dev0",
+  "_name_or_path": "flux-dev",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 16,
+  "latents_mean": null,
+  "latents_std": null,
+  "layers_per_block": 2,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 1024,
+  "scaling_factor": 0.3611,
+  "shift_factor": 0.1159,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": false,
+  "use_quant_conv": false
+}

vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5b59a26851551b67ae1fe58d32e76486e1e812def4696a4bea97f16604d40a3
+size 167666902