Arena AI
A third-party crowdsourced AI model evaluation platform that assesses the real-world performance of frontier AI models through multi-category comparative evaluations.
Timeline (last 90 days)
On the Arena AI (LMArena) code and web development category leaderboard, Kimi K3.1 Pro ranks 8th and is one of only two open-source models in the top 11.
Arena AI (LMArena, formerly Chatbot Arena) uses an anonymous battle format in which users see two models' answers side by side and vote for the better one, which are then aggregated into rankings via an Elo-like algorithm.
In the Arena AI blind-test ELO ranking (coding and web development), Mistral Large 4 ranks 45th, with performance roughly equivalent to Opus 4.8, Sonnet 4.6, or GPT-5.6 level.
On the third-party crowdsourced evaluation platform Arena AI, the model ranks first in overall performance across all 29 categories, ahead of Anthropic's Opus series and some Meta models.
Gemini 4 Argon High ranks first across all 29 categories in the overall Arena AI rankings, surpassing Opus, Astra, and some Meta models.
All Facts (2)
On the Arena AI (LMArena) code and web development category leaderboard, Kimi K3.1 Pro ranks 8th and is one of only two open-source models in the top 11.
50%UnverifiedArena AI (LMArena, formerly Chatbot Arena) uses an anonymous battle format in which users see two models' answers side by side and vote for the better one, which are then aggregated into rankings via an Elo-like algorithm.
50%