Wow! People are already hacking with Llama-3 & making it even better AbacusAI just released Llama-3-Smaug-8b! The model significantly improves the base model's MT-bench score!! MT-bench Scores 1st turn
llama3-smaug-8b 8.79375
llama3-8b-instruct 8.53125 2nd turn

→ View original post on X — @akshay_pachaar