v2 was just released

#1
by LeePapa - opened

https://huggingface.co/Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-v2

Your 27B quant is perfectly compatible with my older V100 GPUs running with vLLM. Any chance you can quant the new version or provide instructions on exactly how I can quant it myself? Also, would be nice to have INT8 as well but I'm at your mercy.

Thanks again!

Sign up or log in to comment