fireworks-ai
/

FLUX.1-schnell-fp8-flumina

Model card Files Files and versions Community

FLUX.1-schnell-fp8-flumina / float8_quantize.py

Commit History

Add fields to configs, fix issue with offload from bnb, remove extra random text code

340f0a0

aredden commited on Aug 19, 2024

Fix for nightly

0aa9861

aredden commited on Aug 18, 2024

cuda version checks

b6617b1

aredden commited on Aug 18, 2024

Fix non-offload inference & add option to load from prequantized flux

2f2c44c

aredden commited on Aug 18, 2024

Add offloading & improved fp8 inference.

28dec30

aredden commited on Aug 18, 2024