arena.ai 推出 Agent Arena Index 智能体排行榜
Check out the NEW Agent Arena Index: https://t.co/sE9q4FSYAt and learn more about the new Coding Ind...
arena.ai 新上了智能体榜单,用数百万真实任务打分。Anthropic 的 Claude Fable 5.2 拿了第一,比 Gemini 4 Carbon 高 134 分。做智能体开发或者选型的人可以直接查各家的细分成绩,比看跑分有参考价值。
arena.ai 发布 Agent Arena Index,作为衡量智能体综合表现的旗舰指标,并新增 Agentic Work、Coding、Chat、Task Outcomes、Tool Use、Steering 六个细分指数,与既有 Alignment Index 一起构成完整评估体系。榜单基于全球社区数百万真实任务的数据打分。
排名方面,Anthropic 的 Claude Fable 5.2 (Max) 以 1463 分居首,领先第二名 134 分;Google DeepMind 的 Gemini 4 Carbon (High) 得 1307 分排第二,OpenAI 的 GPT-6 Astra (Max) 以 1274 分列第三,Meta 的 Muse Wonder Full (Max) 和阿里 Qwen4-Max (xHigh) 分列四五位。
Agent Arena Index 综合四项信号:任务完成的用户确认、赞美与投诉比例、Bash 终端工具使用可靠性,以及对用户纠正的响应能力(Steerability)。Alignment Index 则通过 Deceptive Completion、Unauthorized Action、False Attribution 三项信号衡量智能体是否忠实执行用户意图。
Check out the NEW Agent Arena Index: https://t.co/sE9q4FSYAt and learn more about the new Coding Ind...
Check out the NEW Agent Arena Index: arena.ai/leaderboard/ag… and learn more about the new Coding Index below x.com/arena/status/2… Arena.ai @arena Today, we’re excited to introduce the Agent Arena Index, our new headline measure of overall agent performance, alongside six new focused indexes: Agentic Work, Coding, Chat, Task Outcomes, Tool Use, and Steering. Together with the Alignment Index, these indexes provide a comprehensive view of how agents perform across real-world tasks, from completing complex workflows to using tools and adapting to user feedback. Arena measures agentic performance from millions of real-world tasks by our global community. With the launch of the new indexes, we are updating our scores to provide a more stable and interpretable way to understand how agents perform in the real world. Claude Fable 5.2 (Max) leads the Agent Arena Index by 134 points with a score of 1463. Top 5 labs on the Agent Arena Index: - #1 @AnthropicAI : With Claude Fable 5.2 (Max) at 1463 pts - #2 @GoogleDeepMind : With Gemini 4 Carbon (High) at 1307 pts - #3 @OpenAI : With GPT-6 Astra (Max) at 1274 pts - #4 @AIatMeta : With Muse Wonder Full (Max) at 1200 pts - #5 @Alibaba_Qwen : With Qwen4-Max (xHigh) at 1191 pts The Agent Arena Index combines four key signals of real-world agent performance: - Task Completion: explicit confirmation of task success from users - Praise vs. Complaint: user satisfaction with agent's outcome - Bash Reliability: how well an agent uses terminal tools - Steerability: how well an agent responds to user corrections Alongside capability, the Alignment Index measures how faithfully agents act in accordance with user intent, based on three key signals: Deceptive Completion, Unauthorized Action, and False Attribution. As agents take on increasingly complex, long-horizon work, measuring their real-world capabilities and alignment is more important than ever. Explore the rankings and see what each index measures in the thread below.
This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Agent Arena Index, Anthropic, Claude, GPT-6.
Source attribution: curated from lmarena.ai (@lmarena_ai). Cross-source corroboration: 11 independent sources have covered this development (including Artificial Analysis, Rowan Cheung, rohanpaul_ai), confirming industry-wide signal strength.