File size: 7,983 Bytes
52baaab
 
 
 
c22e483
52baaab
 
2266161
7cae33f
 
9e8dd87
 
84ccd75
f2a58b9
 
84ccd75
 
68d414b
 
f062324
68d414b
 
f062324
68d414b
 
 
 
 
f062324
4acff3b
9b4f8e6
 
5bca1b8
9b4f8e6
d237588
 
5bca1b8
 
f807ff6
d237588
d350f56
 
 
 
 
8cbc120
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c724a07
79a04cc
c724a07
 
 
3dc5d63
 
c724a07
79a04cc
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d350f56
 
 
369d5ab
 
190d547
d66c08a
 
 
 
7cae33f
 
cac27a3
 
190d547
 
cac27a3
 
0523cf8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
---
license: apache-2.0
---

# Repos
https://github.com/mit-han-lab/deepcompressor

# Installation
https://github.com/mit-han-lab/deepcompressor/issues/56

https://github.com/nunchaku-tech/deepcompressor/issues/80

# Windows
https://learn.microsoft.com/en-us/windows/wsl/install

https://www.anaconda.com/docs/getting-started/miniconda/install

# Environment
python 3.10

cuda 12.8

torch 2.7

# Quantization

https://github.com/nunchaku-tech/deepcompressor/blob/main/examples/diffusion/README.md

Model Path: https://github.com/nunchaku-tech/deepcompressor/issues/70#issuecomment-2788155233

Save model: `--save-model true` or `--save-model /PATH/TO/CHECKPOINT/DIR`

Example: `python -m deepcompressor.app.diffusion.ptq examples/diffusion/configs/model/flux.1-dev.yaml examples/diffusion/configs/svdquant/nvfp4.yaml`

Folder Structure 

- refer [black-forest-labs/FLUX.1-dev](https://huggingface.co/black-forest-labs/FLUX.1-dev)

- refer [black-forest-labs/FLUX.1-Kontext-dev](https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev/tree/main)

---

# Blockers
1) NotImplementedError: Cannot copy out of meta tensor; no data! Please use torch.nn.Module.to_empty() instead of torch.nn.Module.to() when moving module from meta to a different device.

potential fix: app.diffusion.pipeline.config.py
```python
    @staticmethod
    def _default_build(
        name: str,
        path: str,
        dtype: str | torch.dtype,
        device: str | torch.device,
        shift_activations: bool
    ) -> DiffusionPipeline:
        if not path:
            if name == "sdxl":
                path = "stabilityai/stable-diffusion-xl-base-1.0"
            elif name == "sdxl-turbo":
                path = "stabilityai/sdxl-turbo"
            elif name == "pixart-sigma":
                path = "PixArt-alpha/PixArt-Sigma-XL-2-1024-MS"
            elif name == "flux.1-dev":
                path = "black-forest-labs/FLUX.1-dev"
            elif name == "flux.1-canny-dev":
                path = "black-forest-labs/FLUX.1-Canny-dev"
            elif name == "flux.1-depth-dev":
                path = "black-forest-labs/FLUX.1-Depth-dev"
            elif name == "flux.1-fill-dev":
                path = "black-forest-labs/FLUX.1-Fill-dev"
            elif name == "flux.1-schnell":
                path = "black-forest-labs/FLUX.1-schnell"
            else:
                raise ValueError(f"Path for {name} is not specified.")
    
        # Instantiate the pipeline
        if name in ["flux.1-canny-dev", "flux.1-depth-dev"]:
            pipeline = FluxControlPipeline.from_pretrained(path, torch_dtype=dtype)
        elif name == "flux.1-fill-dev":
            pipeline = FluxFillPipeline.from_pretrained(path, torch_dtype=dtype)
        elif name.startswith("sana-"):
            if dtype == torch.bfloat16:
                pipeline = SanaPipeline.from_pretrained(
                    path, variant="bf16", torch_dtype=dtype, use_safetensors=True
                )
                pipeline.vae.to(dtype)
                pipeline.text_encoder.to(dtype)
            else:
                pipeline = SanaPipeline.from_pretrained(path, torch_dtype=dtype)
        else:
            pipeline = AutoPipelineForText2Image.from_pretrained(path, torch_dtype=dtype)
    
        # Debug output
        print(">>> DEVICE:", device)
        print(">>> PIPELINE TYPE:", type(pipeline))
    
        # Try to move each component using .to_empty()
        for name in ["unet", "transformer", "vae", "text_encoder"]:
            module = getattr(pipeline, name, None)
            if isinstance(module, torch.nn.Module):
                try:
                    print(f">>> Moving {name} to {device} using to_empty()")
                    module.to_empty(device)
                except Exception as e:
                    print(f">>> WARNING: {name}.to_empty({device}) failed: {e}")
                    try:
                        print(f">>> Falling back to {name}.to({device})")
                        module.to(device)
                    except Exception as ee:
                        print(f">>> ERROR: {name}.to({device}) also failed: {ee}")
    
        # Identify main model (for patching)
        model = getattr(pipeline, "unet", None) or getattr(pipeline, "transformer", None)
        if model is not None:
            replace_fused_linear_with_concat_linear(model)
            replace_up_block_conv_with_concat_conv(model)
            if shift_activations:
                shift_input_activations(model)
        else:
            print(">>> WARNING: No model (unet/transformer) found for patching")
    
        return pipeline
```

Debug Log
```
25-07-21 22:47:02 | I | === Start Evaluating ===
25-07-21 22:47:02 | I | * Building diffusion model pipeline
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 15.44it/s]
Loading pipeline components...:  57%|████████████████████████████████████████████████████████████████                                            | 4/7 [00:00<00:00,  9.47it/s]
You set `add_prefix_space`. The tokenizer needs to be converted from the slow tokenizers
Loading pipeline components...: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:00<00:00,  7.79it/s]
>>> DEVICE: cuda
>>> PIPELINE TYPE: <class 'diffusers.pipelines.flux.pipeline_flux.FluxPipeline'>
>>> Moving transformer to cuda using to_empty()
>>> WARNING: transformer.to_empty(cuda) failed: Module.to_empty() takes 1 positional argument but 2 were given
>>> Falling back to transformer.to(cuda)
>>> ERROR: transformer.to(cuda) also failed: Cannot copy out of meta tensor; no data! Please use torch.nn.Module.to_empty() instead of torch.nn.Module.to() when moving module from meta to a different device.
>>> Moving vae to cuda using to_empty()
>>> WARNING: vae.to_empty(cuda) failed: Module.to_empty() takes 1 positional argument but 2 were given
>>> Falling back to vae.to(cuda)
>>> Moving text_encoder to cuda using to_empty()
>>> WARNING: text_encoder.to_empty(cuda) failed: Module.to_empty() takes 1 positional argument but 2 were given
>>> Falling back to text_encoder.to(cuda)
25-07-21 22:47:05 | I |   Replacing fused Linear with ConcatLinear.
25-07-21 22:47:05 | I |     + Replacing fused Linear in single_transformer_blocks.0 with ConcatLinear.
25-07-21 22:47:05 | I |       - in_features = 3072/15360
25-07-21 22:47:05 | I |       - out_features = 3072
25-07-21 22:47:05 | I |     + Replacing fused Linear in single_transformer_blocks.1 with ConcatLinear.
25-07-21 22:47:05 | I |       - in_features = 3072/15360
25-07-21 22:47:05 | I |       - out_features = 3072
25-07-21 22:47:05 | I |     + Replacing fused Linear in single_transformer_blocks.2 with ConcatLinear.
25-07-21 22:47:05 | I |       - in_features = 3072/15360
25-07-21 22:47:05 | I |       - out_features = 3072
```

2) KeyError: <class 'diffusers.models.transformers.transformer_flux.FluxAttention'>


---

# Dependencies
https://github.com/Dao-AILab/flash-attention

https://github.com/facebookresearch/xformers

https://github.com/openai/CLIP

https://github.com/THUDM/ImageReward

# Wheels

https://huggingface.co/datasets/siraxe/PrecompiledWheels_Torch-2.8-cu128-cp312

https://huggingface.co/lldacing/flash-attention-windows-wheel