Flux.1_collection / README.md
marduk191's picture
Update README.md
d9766db verified
metadata
tags:
  - not-for-all-audiences

Unified single file versions of flux.1 for comfyui. All files are set to upcast to bf16 and have a baked in VAE and clip L included:

flux.1_dev_8x8_e4m3fn-marduk191.safetensors is Flux.1 Dev quantized to 8 bit with an 8 bit T5 XXL encoder included.

flux.1_dev_fp8_fp16t5-marduk191.safetensors is Flux.1 Dev quantized to 8 bit with an 16 bit T5 XXL encoder included.

flux.1_schnell_8x8_e4m3fn-marduk191.safetensors is Flux.1 Schnell quantized to 8 bit with an 8 bit T5 XXL encoder included.

flux.1_schnell_fp8_fp16t5-marduk191.safetensors is Flux.1 Schnell quantized to 8 bit with an 16 bit T5 XXL encoder included.

flux.1_dev_16x16-marduk191.safetensors Flux.1 Dev quantized to 16 bit with an 16 bit T5 XXL encoder included.

flux.1_schnell_16x16-marduk191.safetensors Flux.1 Schnell quantized to 16 bit with an 16 bit T5 XXL encoder included.

8 bit model with a 16 bit text encoder is a decent compromise for speed with a slightly reduced image quality.

Workflow examples are available here: https://civitai.com/models/619254/marduk191s-flux1-schnell-workflow

Discord

ko-fi