Omnigen V1 Fp16 Bnb 4bit

🚀 OmniGen-v1 4bit-NF4 float16 Model

This repository provides the 4bit-NF4 float16 model weights of OmniGen-v1 using bitsandbytes. These weights are designed for Google Colab users or those with GPUs that do not support bfloat16. For other 4-bit users, the bf16-bnb-4bit model is recommended as it can generate higher-quality images. For more information about OmniGen, please refer to the original model card.

📄 License

This project is licensed under the MIT license.

📦 Model Information

🚀 Quick Start

Before you start, set up your environment by following the original Quick Start Guide.

⚠️ Important Note

This feature is not officially supported yet. You'll need to install the repo from this pull request.

💻 Usage Examples

Basic Usage

from OmniGen import OmniGenPipeline, OmniGen
import torch

# pass the quantized model in the pipeline
model = OmniGen.from_pretrained('gryan/OmniGen-v1-fp16-bnb-4bit', dtype=torch.float16)
pipe = OmniGenPipeline.from_pretrained("Shitao/OmniGen-v1", model=model)

# proceed as normal!

## Text to Image
images = pipe(
    prompt="A curly-haired man in a red shirt is drinking tea.", 
    height=1024, 
    width=1024, 
    guidance_scale=2.5,
    seed=0,
)
images[0].save("example_t2i.png")  # save output PIL Image

## Multi-modal to Image
# In the prompt, we use the placeholder to represent the image. The image placeholder should be in the format of <img><|image_*|></img>
# You can add multiple images in the input_images. Please ensure that each image has its placeholder. For example, for the list input_images [img1_path, img2_path], the prompt needs to have two placeholders: <img><|image_1|></img>, <img><|image_2|></img>.
images = pipe(
    prompt="A man in a black shirt is reading a book. The man is the right man in <img><|image_1|></img>.",
    input_images=["./imgs/test_cases/two_man.jpg"],
    height=1024, 
    width=1024,
    guidance_scale=2.5, 
    img_guidance_scale=1.6,
    seed=0
)
images[0].save("example_ti2i.png")  # save output PIL image

🖼️ Image Samples

🔗 Other Model Versions

• 8-bit weights: gryan/OmniGen-v1-bnb-8bit
• 4-bit (bf16, nf4) weights: gryan/OmniGen-v1-bnb-4bit