---
license: apache-2.0
language:
- zh
- en
base_model:
- THUDM/glm-4-9b
pipeline_tag: text-to-image
library_name: diffusers
---
# CogView4-6B
🤗 Space |
🌐 Github |
📜 arxiv
## Inference Requirements and Model Introduction
+ Resolution: Width and height must be between `512px` and `2048px`, divisible by `32`, and ensure the maximum number of pixels does not exceed `2^21` px.
+ Precision: BF16 / FP32 (FP16 is not supported as it will cause overflow resulting in completely black images)
Using `BF16` precision with `batchsize=4` for testing, the memory usage is shown in the table below:
| Resolution | enable_model_cpu_offload OFF | enable_model_cpu_offload ON | enable_model_cpu_offload ON Text Encoder 4bit |
|--------------|------------------------------|-----------------------------|----------------------------------------------------|
| 512 * 512 | 33GB | 20GB | 13G |
| 1280 * 720 | 35GB | 20GB | 13G |
| 1024 * 1024 | 35GB | 20GB | 13G |
| 1920 * 1280 | 39GB | 20GB | 14G |
| 2048 * 2048 | 43GB | 21GB | 14G |
## Quick Start
First, ensure you install the `diffusers` library from source.
```shell
pip install git+https://github.com/huggingface/diffusers.git
cd diffusers
pip install -e .
```
Then, run the following code:
```python
from diffusers import CogView4Pipeline
pipe = CogView4Pipeline.from_pretrained("THUDM/CogView4-6B", torch_dtype=torch.bfloat16)
# Open it for reduce GPU memory usage
pipe.enable_model_cpu_offload()
pipe.vae.enable_slicing()
pipe.vae.enable_tiling()
prompt = "A vibrant cherry red sports car sits proudly under the gleaming sun, its polished exterior smooth and flawless, casting a mirror-like reflection. The car features a low, aerodynamic body, angular headlights that gaze forward like predatory eyes, and a set of black, high-gloss racing rims that contrast starkly with the red. A subtle hint of chrome embellishes the grille and exhaust, while the tinted windows suggest a luxurious and private interior. The scene conveys a sense of speed and elegance, the car appearing as if it's about to burst into a sprint along a coastal road, with the ocean's azure waves crashing in the background."
image = pipe(
prompt=prompt,
guidance_scale=3.5,
num_images_per_prompt=1,
num_inference_steps=50,
width=1024,
height=1024,
).images[0]
image.save("cogview4.png")
```
## Model Performance
We've tested on multiple benchmarks and achieved the following scores:
### dpg_bench
| model | overall | global | entity | attribute | relation | other |
|-------|---------|--------|--------|-----------|----------|-------|
| sdxl | 74.65 | 83.27 | 82.43 | 80.91 | 86.76 | 80.41 |
| pixart-alpha | 71.11 | 74.97 | 79.32 | 78.60 | 82.57 | 76.96 |
| sd3-medium | 84.08 | 87.90 | **91.01** | 88.83 | 80.70 | 88.68 |
| dalle-3 | 83.50 | **90.97** | 89.61 | 88.39 | 90.58 | 89.83 |
| flux.1-dev | 83.79 | 85.80 | 86.79 | 89.98 | 90.04 | **89.90** |
| **cogview4** | **85.13** | 83.85 | 90.35 | **91.17** | **91.14** | 87.29 |
### Geneval
| model | overall | single | two | counting | colors | position | Color attribution |
|-------|---------|--------|-----|----------|--------|----------|------------------|
| sdxl | 0.55 | 0.98 | 0.74 | 0.39 | 0.85 | 0.15 | 0.23 |
| pixart-alpha | 0.48 | 0.98 | 0.50 | 0.44 | 0.80 | 0.08 | 0.07 |
| sd3-meidum | **0.74** | **0.99** | **0.94** | **0.72** | **0.89** | 0.33 | **0.60** |
| dall-e 3 | 0.67 | 0.96 | 0.87 | 0.47 | 0.83 | 0.43 | 0.45 |
| flux.1-dev | 0.66 | 0.98 | 0.79 | 0.73 | 0.77 | 0.22 | 0.45 |
| **cogview4** | 0.73 | **0.99** | 0.86 | 0.66 | 0.79 | **0.48** | 0.58 |
### t2i_compbench
| model | color | shape | texture | 2d-spatial | 3d-spatial | numeracy | Non-spatial clip | complex 3-in-1 |
|-------|-------|-------|---------|------------|------------|----------|-----------------|---------------|
| sdxl | 0.5879 | 0.4687 | 0.5299 | 0.2133 | 0.3566 | 0.4988 | 0.3119 | 0.3237 |
| pixart-alpha | 0.6690 | 0.4927 | 0.6477 | 0.2064 | 0.3901 | 0.5058 | **0.3197** | 0.3433 |
| sd3-medium | **0.8132** | 0.5885 | **0.7334** | **0.3200** | **0.4084** | 0.6174 | 0.3140 | 0.3771 |
| dall-e 3 | 0.7785 | **0.6205** | 0.7036 | 0.2865 | 0.3744 | 0.5880 | 0.3003 | **0.3773** |
| flux.1-dev | 0.7572 | 0.5066 | 0.6300 | 0.2700 | 0.3992 | 0.6165 | 0.3065 | 0.3628 |
| **cogview4** | 0.7786 | 0.5880 | 0.6983 | 0.3075 | 0.3708 | **0.6626** | 0.3056 | 0.3869 |
## Chinese Text Accuracy Evaluation
| model | Precision | Recall | F1 Score | pick@4 |
|-------|-----------|--------|----------|--------|
| kolors | 0.6094 | 0.1886 | 0.2880 | 0.1633 |
| **cogview4** | **0.6969** | **0.5532** | **0.6168** | **0.3265** |
## Citation
🌟 If you find our work helpful, please consider citing our paper and leaving valuable stars
```
@article{zheng2024cogview3,
title={Cogview3: Finer and faster text-to-image generation via relay diffusion},
author={Zheng, Wendi and Teng, Jiayan and Yang, Zhuoyi and Wang, Weihan and Chen, Jidong and Gu, Xiaotao and Dong, Yuxiao and Ding, Ming and Tang, Jie},
journal={arXiv preprint arXiv:2403.05121},
year={2024}
}
```
## License
This model is released under the [Apache 2.0 License](LICENSE).
