Model accuracy

#12
by Qnibbles - opened

Are there some benchmarks for accuracy/loss compared to the FP16 or FP8 version, like https://huggingface.co/RedHatAI/Qwen3.5-122B-A10B-NVFP4 for example?
Thanks for the quant.

not that im aware off but the redhaiAI version was incredible bad compaired to this version . I have yet found a 122B version better than this. if anyone has discovered one let me know.

Sign up or log in to comment