#livecodebenchpro — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #livecodebenchpro, aggregated by home.social.
-
Google just rolled out Gemini 3.1 Pro, smashing the GPQA Diamond benchmark at 94.3% and climbing to an Elo 2 on LiveCodeBench Pro. It also tops SWE‑Bench, showing leaps in AI reasoning, scientific knowledge, and vibe‑coding. Curious how it reshapes open‑source AI research? Read the full breakdown. #Gemini3_1Pro #GPQADiamond #LiveCodeBenchPro #SWEBench
🔗 https://aidailypost.com/news/google-unveils-gemini-31-pro-hits-943-gpqa-diamond-coding-elo-2