We got 207 tok/s with Qwen3.5-27B on an RTX 3090

Lucebox optimization hub: hand-tuned LLM inference, built for specific consumer hardware. - Luce-Org/lucebox-hub

148 points by GreenGames 184 days ago github.com 42 comments

Summary not yet generated for this article.

Loading more comments... ⌛

Scroll down to load more comments

No comments available.