Drop in replacement for HF's TGI server. The fastest and easiest way to inference LLMs locally<p>Github: <a href="https://github.com/titanml/takeoff">https://github.com/titanml/takeoff</a>
Docs: <a href="https://docs.titanml.co/docs/titan-takeoff/getting-started" rel="nofollow noreferrer">https://docs.titanml.co/docs/titan-takeoff/getting-started</a>
Discord: <a href="https://discord.gg/83RmHTjZgf" rel="nofollow noreferrer">https://discord.gg/83RmHTjZgf</a>