image-text-to-text
mit
NVFP4FP8
What this is
Takes pictures and text together and answers in text. You can hand it a screenshot and ask about it.
Packaged as NVFP4, FP8. NVFP4 is 4-bit and needs a Blackwell GPU.
mit — commercial use allowed.
Release chainonly the steps this archive can prove
- weights posted2026-09-16
Measured herefrom our own daily capture — upstream publishes today only
← the board