When Meta, the guardian firm of Fb, introduced its newest open-source giant language mannequin (LLM) on July twenty third, it claimed that probably the most highly effective model of Llama 3.1 had “state-of-the-art capabilities that rival the most effective closed-source fashions” equivalent to GPT-4o and Claude 3.5 Sonnet. Meta’s announcement included a desk, exhibiting the scores achieved by these and different fashions on a sequence of well-liked benchmarks with names equivalent to MMLU, GSM8Okay and GPQA.
