OpenAI Launches GPT-5, Claims State-of-the-Art Results Across the Board
GPT-5 hits 94.6% on AIME 2025 and 74.9% on SWE-bench Verified, cuts hallucinations sharply, and replaces five prior models as ChatGPT's default.
Topic
Topic
Topic
GPT-5 hits 94.6% on AIME 2025 and 74.9% on SWE-bench Verified, cuts hallucinations sharply, and replaces five prior models as ChatGPT's default.