Agent Arena 用真实任务数据解决了智能体评估难的问题，做 AI 智能体开发或选型的团队可以直接参考排行榜和会话信号，比纯基准测试更有说服力。
Arena.ai 宣布将 NVIDIA 的 Nemotron 3 Ultra 模型加入其新的 Agent Mode，用于评估 AI 智能体在真实世界中的多步骤任务能力。Agent Arena 通过数百万次真实用户会话，测量模型在编写代码、创建演示文稿、网络研究、构建应用和文档分析等复杂工作流中的表现。评估基于任务成功率、可操控性、错误恢复、用户反馈和工具幻觉五个信号。当前排行榜显示 OpenAI GPT-5.5 排名第一，Anthropic Claude-Opus-4.7 第二，Zai GLM-5.1 第三。该平台已积累超过 30 万任务、200 万次工具调用和 4000 万行代码数据。
Nemotron 3 Ultra has been added to the new Agent Mode! This latest model from @NVIDIA and other to...
Nemotron 3 Ultra has been added to the new Agent Mode! This latest model from @NVIDIA and other top frontier models are ready for your complex, multi-step tasks. Your sessions will help shape the new Agent Arena leaderboard. Arena.ai @arena Introducing Agent Arena: real-world agentic evals at scale. How do you evaluate agents doing actual work? We measure millions of live sessions where real users accomplish real tasks. On Arena, models now get web search, filesystem, and terminal tools to complete complex workflows: writing code, creating slide deck, researching the web, building apps, and analyzing documents. Every session produces rich signals. Users iterate with the agent turn-by-turn: approving, editing, correcting, praise or expressing frustration. The environment gives feedback too: shell errors, tool failures, recovery attempts, and more. Our leaderboard measures each model's agentic performance using causal inference across five signals: task success, steerability, error recovery, user praise vs. complaint, and tool hallucination. This leaderboard snapshot is built from 300K+ tasks, 2M+ tool calls, and 40M lines of code by agents. Top labs in Agent Arena: - #1 @OpenAI : GPT-5.5 (High) - #2 @AnthropicAI : Claude-Opus-4.7 (Thinking) - #3 @Zai_org : GLM-5.1 - #4 @GoogleDeepMind : Gemini-3.1-Pro - #5 @Kimi_Moonshot : Kimi-K2.6 More analysis in the thread, with the full technical blog below. 🔗 View Quoted Tweet 💬 1 🔄 0 ❤️ 16 👀 1356 📊 2 ⚡