InstantX/flux-dev-de-distill-diffusers

1---2license: other3license_name: flux-1-dev-non-commercial-license4license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md5language:6- en7base_model:8- black-forest-labs/FLUX.1-dev9library_name: diffusers10---11 12## Model Details13This is a diffusers version of [nyanko7/flux-dev-de-distill](https://huggingface.co/nyanko7/flux-dev-de-distill) that generates exactly the same results as the original. But please note that there is a certain degree of quality degradation in comparison with [black-forest-labs/FLUX.1-dev](https://huggingface.co/black-forest-labs/FLUX.1-dev) in our test. 14We are training its fine-tuning models to check its effectiveness, feel free to let us know if you have any interesting findings.15 16## Usage17```python18import torch19from diffusers import FluxTransformer2DModel20from pipeline_flux_de_distill import FluxPipeline21 22model_path = "black-forest-labs/FLUX.1-dev"23    24transformer = FluxTransformer2DModel.from_pretrained(25    "InstantX/flux-dev-de-distill-diffusers",26    torch_dtype=torch.bfloat1627)28 29pipeline = FluxPipeline.from_pretrained(model_path, transformer=transformer, torch_dtype=torch.bfloat16).to("cuda")30 31prompt = "a tiny astronaut hatching from an egg on the moon"32negative_prompt = "bad photo"33 34image = pipeline(35    prompt=prompt,36    negative_prompt=negative_prompt,37    guidance_scale=3.5,38    num_inference_steps=24,39).images[0]40 41image.save("de-distill.jpg")42```43 44## Other Resources45- [MinusZoneAI/flux-dev-de-distill-fp8](https://huggingface.co/MinusZoneAI/flux-dev-de-distill-fp8), [TheYuriLover/flux-dev-de-distill-GGUF](https://huggingface.co/TheYuriLover/flux-dev-de-distill-GGUF)46- [ashen0209/Flux-Dev2Pro](https://huggingface.co/ashen0209/Flux-Dev2Pro), [ostris/OpenFLUX.1](https://huggingface.co/ostris/OpenFLUX.1)47 48 49## Acknowledgements50This project is co-sponsored by [HuggingFace](https://huggingface.co/) and [fal](https://huggingface.co/fal).