This model is a scam. Benchmarks are actually Sonnet 3.5 with a system prompt. Actual model performance is somehow WORSE than original Llama 3.1.
· Sign up or log in to comment
