Henry Garner @hendroid.io · Dec 9

Shiny! The newly released Llama 3.3 LLM leads the LiveBench ranking for instruction following¹, beating Claude 3.5, GPT-4o, OpenAI o1, and you can run it on your local² machine. > ollama run llama3.3 livebench.ai#/?IF=as

7 likes 3 replies

?

Replies

Clement Rousseau · Dec 10

Crazy, I just read this morning Gemini was on top of the chart. I guess it didn't last long 😀

Nathanael · Dec 9

Wow, wasn’t expecting open models to catch up this quickly. This is awesome!

Henry Garner · Dec 9

¹ Tasks which paraphrase, simplify, summarise or generate stories from news articles ² Provided you have 64GB RAM…