Claude 3.7 Dominates Mario for 90 Seconds, GPT-4o Fails Immediately! Karpathy Declares Benchmark Invalidated, Gaming as a New Frontier for LLM Evaluation

⌘K

F

Claude 3.7 Dominates Mario for 90 Seconds, GPT-4o Fails Immediately! Karpathy Declares Benchmark Invalidated, Gaming as a New Frontier for LLM Evaluation | BestBlogs.dev