Fast, few-step text-to-image generation.
Derived from black-forest-labs/FLUX.1-schnell (Apache-2.0); repackaged by Zen.
This repository repackages Black Forest Labs FLUX.1-schnell (FluxPipeline, dual CLIP+T5 text encoders). No weights were trained from scratch by Zen.
Developed/repackaged by Hanzo AI and the Zoo Labs Foundation.
from diffusers import FluxPipeline
import torch
pipe = FluxPipeline.from_pretrained("zenlm/zen3-image-fast", torch_dtype=torch.bfloat16)
pipe = pipe.to("cuda")
image = pipe("A serene mountain landscape at sunset", num_inference_steps=4, guidance_scale=0.0).images[0]
image.save("output.png")
| Attribute | Value |
|---|---|
| Base model | black-forest-labs/FLUX.1-schnell |
| Pipeline | FluxPipeline |
| License | Apache 2.0 |
Apache 2.0. This is a derivative repackaging of FLUX.1-schnell by Black Forest Labs, released under the Apache License 2.0. All rights and attribution for the underlying model belong to the original authors.
This product includes the FLUX.1-schnell model developed by Black Forest Labs, licensed under the Apache License 2.0.
transformer/diffusion_pytorch_model-00001-of-00003.safetensors 10.0 GBtransformer/diffusion_pytorch_model-00002-of-00003.safetensors 9.9 GBtext_encoder_2/model-00001-of-00002.safetensors 5.0 GBtext_encoder_2/model-00002-of-00002.safetensors 4.5 GBtransformer/diffusion_pytorch_model-00003-of-00003.safetensors 3.9 GBae.safetensors 335 MBtext_encoder/model.safetensors 246 MBvae/diffusion_pytorch_model.safetensors 168 MBblack-forest-labs/FLUX.1-schnell