Llama 3.3 70B Open-weight 70B model. Benchmark results Service providers HumanEval 88.4% MMLU 86.0% MATH 77.0% No serving provider data recorded yet.
Llama 3.3 70B Llama 3.3 70B