Flux.1-Dev Hyper NF4 + Flux.1-Dev BNB NF4 + Flux.1-Schnell BNB NF4
Flux.1-Dev Hyper NF4:
Source: https://huggingface.co/ZhenyaYang/flux_1_dev_hyper_8steps_nf4/tree/main from ZhenyaYang converted Hyper-SD to NF4 (8 steps)
Flux.1-Dev BNB NF4 (v1 & v2):
Source: https://huggingface.co/lllyasviel/flux1-dev-bnb-nf4/tree/main from lllyasviel
Flux.1-Schnell BNB NF4:
Source: https://huggingface.co/silveroxides/flux1-nf4-weights/tree/main from silveroxides
ComfyUI: https://github.com/comfyanonymous/ComfyUI_bitsandbytes_NF4
Forge: https://github.com/lllyasviel/stable-diffusion-webui-forge/discussions/981
☕ Buy me a coffee: https://ko-fi.com/ralfingerai
🍺 Join my discord: https://discord.gg/g5Pb8qNUuP
Flux.1-Dev Hyper NF4 + Flux.1-Dev BNB NF4 + Flux.1-Schnell BNB NF4 Flux.1-Dev BNB NF4 v2
V2 is 0.5 GB larger than the previous version, since the chunk 64 norm is now stored in full precision float32, making it much more precise than the previous version. Also, since V2 does not have second compression stage, it now has less computation overhead for on-the-fly decompression, making the inference a bit faster!
Flux.1-Dev Hyper NF4 + Flux.1-Dev BNB NF4 + Flux.1-Schnell BNB NF4 Flux.1-Dev BNB NF4 v1
Main model in bnb-nf4
T5xxl in fp8e4m3fn
CLIP-L in fp16
VAE in bf16