File size: 2,862 Bytes
dbb180b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model:
  - black-forest-labs/FLUX.1-dev
base_model_relation: quantized
pipeline_tag: text-to-image
tags:
- dfloat11
- df11
- lossless compression
- 70% size, 100% accuracy
---

## DFloat11 Compressed Model: `black-forest-labs/FLUX.1-dev`

This is a **losslessly compressed** version of [`black-forest-labs/FLUX.1-dev`](https://huggingface.co/black-forest-labs/FLUX.1-dev) using our custom **DFloat11** format. The outputs of this compressed model are **bit-for-bit identical** to the original BFloat16 model, while reducing GPU memory consumption by approximately **30%**.

### 🔍 How It Works

DFloat11 compresses model weights using **Huffman coding** of BFloat16 exponent bits, combined with **hardware-aware algorithmic designs** that enable efficient on-the-fly decompression directly on the GPU. During inference, the weights remain compressed in GPU memory and are **decompressed just before matrix multiplications**, then **immediately discarded after use** to minimize memory footprint.

Key benefits:

* **No CPU decompression or host-device data transfer** --- all operations are handled entirely on the GPU.
* DFloat11 is **much faster than CPU-offloading approaches**, enabling practical deployment in memory-constrained environments.
* The compression is **fully lossless**, guaranteeing that the model’s outputs are **bit-for-bit identical** to those of the original model.

### 🔧 How to Use

1. Install the DFloat11 pip package *(installs the CUDA kernel automatically; requires a CUDA-compatible GPU and PyTorch installed)*:

    ```bash
    pip install dfloat11[cuda12]
    # or if you have CUDA version 11:
    # pip install dfloat11[cuda11]
    ```

2. To use the DFloat11 model, run the following example code in Python:
    ```python
    import torch
    from diffusers import FluxPipeline
    from dfloat11 import DFloat11Model

    pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16)
    pipe.enable_model_cpu_offload()

    DFloat11Model.from_pretrained('DFloat11/FLUX.1-dev-DF11', device='cpu', bfloat16_model=pipe.transformer)

    prompt = "A futuristic cityscape at sunset, with flying cars, neon lights, and reflective water canals"
    image = pipe(
        prompt,
        width=1920,
        height=1440,
        guidance_scale=3.5,
        num_inference_steps=50,
        max_sequence_length=512,
        generator=torch.Generator(device="cuda").manual_seed(0)
    ).images[0]

    image.save("image.png")
    ```

### 📄 Learn More

* **Paper**: [70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length Float](https://arxiv.org/abs/2504.11651)
* **GitHub**: [https://github.com/LeanModels/DFloat11](https://github.com/LeanModels/DFloat11)
* **HuggingFace**: [https://huggingface.co/DFloat11](https://huggingface.co/DFloat11)