Updated: May 16, 2025
base modelDownload
1 variant available
bf16 SafeTensor
speedfp8_e5m2.safetensors
BF16, good balance (pruned) β’ 15.9 GB
Verified: 2 years ago
2,3080 1 2 3 4 5 6 7 8 9,0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
(124)
Aug 8, 2024
Flux.1 S
I found little enhancment using e5m2 quantization !

1K0 1 2 3 4 5 6 7 8 9K
2.9K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K

License:
Apache 2.0β¨ What Makes This Model So Special? β¨
π Perfect fusion of Flux1-dev and Flux1-schnell strengths
πΎ Full FP8 version (includes CLIP and VAE)
π» Runs smoothly on modest 12GB GPUs
πΌοΈ Premium quality outputs rivaling resource-hungry monsters
π οΈ How To Get The Most Out Of It:
π Advanced Workflow:
π Setup Structure
π ComfyUI/
βββ π models/
β βββ πcheckpoints/
β β βββ SPEED-FP8.safetensors
π» Tested on forge https://civitai.com/images/24458443
π Special Thanks
Big thanks to @Dunc4n1dah0 ! π

π¨βπ» Developer Information
This workflow guide was created by Abdallah Al-Swaiti:
For additional tools and updates, check out the OllamaGemini Node: GitHub Repository

