Arena Launches the Agent Arena Index and Six Specialized Agent Indices
Arena has launched the Agent Arena Index to measure agents’ overall performance, along with six specialized indices: Agentic Work, Coding, Chat, Task Outcomes, Tool Use, and Steering. Drawing on millions of real-world tasks from its global community, the index combines four categories of signals: task completion, positive and negative user feedback, Bash reliability, and steerability. Claude Fable 5.2 (Max) ranks first with a score of 1463. The accompanying Alignment Index assesses whether agents follow user intent based on deceptive completion, unauthorized actions, and misattribution.
Why read this
AI score 62
Based on millions of real-world tasks and incorporating four categories of performance signals, the index provides a more stable framework for comparing agent capabilities and alignment.
Topics
View original posthttps://t.co/YieG0GYBat
View original
https://t.co/YieG0GYBat
Body
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.
AI Trends