flux-3-image-klein-FP8

This is an FP8 / INT8 quantized version of FLUX 3 Image Klein (the 16-step distilled checkpoint). - Optimized for efficient inference with reduced memory footprint: the transformer is 6.44 GB instead of 11.1 GB. Same-seed LPIPS vs the bf16 model (lower is better): 0.035 INT8, 0.036 FP8. - Loads pre-quantized in Unsloth Desktop, which skips the bf16 download and the on-the-fly quantization. The text encoder (Qwen3-VL-4B-Instruct) and the VAE are unchanged; take them from black-forest-labs/flux-3-image-klein. The text encoder rows are scored against the bf16 text encoder with the bf16 transformer, so they measure only the encoder. The VAE stays bf16. This repo also carries the upstream diffusers folders from black-forest-labs/flux-3-image-klein, unchanged (`modular_model_index.json`, `text_encoder/`, `transformer/`, `vae/`), so Unsloth Desktop downloads every component from here.

번역 준비 중입니다. 원문을 표시합니다.