People who tried gpu-unlock on 3090s and such - does this make any difference on mixed-precision quants where you get some parts of the model quantized to <=Q8 but keep some tensors at BF16?

1. 

   消息推送

   People who tried gpu-unlock on 3090s and such - does this make any difference on mixed-precision quants where you get some parts of the model quantized to <=Q8 but keep some tensors at BF16?

   3小时前

   People who tried gpu-unlock on 3090s and such - does this make any difference on mixed-precision quants where you get some parts of the model quantized to <=Q8 but keep some tensors at BF16?
   
   🌐 链接: https://www.reddit.com/r/LocalLLaMA/comments/1xg0kle/people_who_tried_gpuunlock_on_3090s_and_such_does/
   🏷️ 分组: Reddit｜AI·自托管·VPS
   🕒 时间: 2026-10-26 23:48:47

   Reddit

   

   From the LocalLLaMA community on Reddit

   Explore this post and more from the LocalLLaMA community

   3小时前