Those results are more useful as a map of where models appear strongest than as a proclamation of one permanent winner — ...

A mysterious model benchmarked on Arena.ai appears to be Google's unreleased Gemini 4 Pro, featuring 10M context windows, ...

Arena CEO Anastasios Angelopoulos said companies are "terrified" of working with frontier AI companies and China-based open models.

I’ve spent a good deal of my career thinking about how we structure and fund research in the biomedical sciences in the U.S. and questioning whether the U.S. model of embedding research in ...

OpenAI dropped its latest image generation model, ChatGPT Images 2.5, on September 8, and it didn’t take long for the results ...

Just eight months after launching its commercial service, AI leaderboard provider Arena, which originated as a research project at UC Berkeley in 2023, has reached $100 million in annualized run-rate ...

Google’s Gemma 4 family just posted a result that will get attention in the open-source AI community: its 31-billion-parameter dense model has climbed to third place among all open models on the Arena ...

OpenAI's new chatgpt-4o-latest model has significantly boosted ChatGPT's performance, leading to OpenAI reclaiming the No. 1 spot in the LMSYS Chatbot Arena with a record score of 1314. Last week, ...