BenchmarkLMARENA排行榜
LMARENA
LMARENA (formerly Chatbot Arena) is an open benchmark platform for evaluating and comparing the capabilities of large language models. Its core mechanism uses human preference voting, presenting users with anonymous blind comparisons of outputs from two models and aggregating real user evaluations into a model leaderboard. The platform is characterized by crowdsourced evaluation, dynamic updates, and coverage of multilingual and multi-task scenarios, and is widely referenced for model performance comparison in academia and industry.
Core Facts
Source Articles
Chinese Open-Source Models Trail US Closed-Source by Only 7 Months: The AI Race Landscape Is ShiftingOct 14Eight AI Models Tested: A Full Record of Dory's Production Environment Model SelectionOct 13LMArena Gets a New CEO: Opportunities and Concerns for an AI Benchmarking PlatformSep 25AA Index Update: Fable 5.1 and GPT-astra Tied at the Top — A Deep Dive into the LLM Competitive LandscapeSep 11Three Major Chinese AI Models Incoming: Qwen 4, DeepSeek V4, and GLM Next Version BreakdownJul 23Gemini 3.5 Pro Leak Analysis: Coding Matches GPT 5.5, Spark Agent Sparks Privacy ControversyJun 3AI Daily: Claude Autonomous Tasks Exceed 16 Hours, GPT 5.5 Proves Mathematical TheoremsJun 3Cursor 2.0 Deep Dive: Hands-On Testing of Five Major Features Including Custom Models and Multi-Agent Parallel DevelopmentJun 2