Comment

Comment by yolo-auto

1x MI300x , rocm/sglang , highly optimized image, FP8, FP16 Cache (this fixes looping issues with qwen3.6-35b

~100tok/s average

parent yolo-auto 3 months ago

No comments yet

Be the first to share your thoughts.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton