Advertisement
Advertisement
TOPIC

LMArena

LMArena
The latest news and top stories on LMArena. A prominent San Francisco Bay Area-based research institute, LMArena, formerly Chatbot Arena, is dedicated to evaluating large language models through anonymous crowd-sourced comparisons. Originating as a UC Berkeley research project, it has evolved into a company, Arena Intelligence Inc., providing a public web-based platform for benchmarking AI models. Its main activities include maintaining a dynamic leaderboard, fostering open-source AI development and offering pre-release testing. LMArena is recognised for its significant contribution to transparent community-driven AI evaluation, attracting millions of users and influencing major AI labs.
Artificial intelligence

Preview of Alibaba’s top AI model leads Chinese peers, lags US rivals

Qwen3.5-Max-Preview enters a global ranking at 15th, behind models from Anthropic, OpenAI and Google.

Tencent’s AI image model beats Google’s Nano Banana in leaderboard

Alibaba’s Qwen3 and Moonshot’s Kimi-K2 storm into global AI rankings

videocam

Chinese open-source AI models occupy top spots among global developers: ranking

Kimi K2, MiniMax M1, Qwen 3 and a variant of DeepSeek R1 rank as the world’s top open-sourced AI models, according to LMArena.

Advertisement
Advertisement
Advertisement
Advertisement
Advertisement
Advertisement