精选理由
Qwen放出了新基准分，PaperBench直接到93，比之前64.8涨了一大截，智能体任务可以关注。
Qwen在长期任务和视觉智能体基准上报告了新成绩。PaperBench从64.8升至93.0，OSWorld-Verified从73.3升至86.1，Vision2Web从42.1升至69.0。数据由OpenRouter推文转述。
原文 · OpenRouter
Against Qwen3.7 Max, Qwen reports gains across long-horizon and visual agent benchmarks: PaperBench ...
Against Qwen3.7 Max, Qwen reports gains across long-horizon and visual agent benchmarks: PaperBench rises from 64.8 to 93.0, OSWorld-Verified from 73.3 to 86.1, and Vision2Web from 42.1 to 69.0. 💬 1 🔄 0 ❤️ 7 👀 1075 📊 2 ⚡