Qwen 团队开源多语言 AI Agent 基准 BabelArena，覆盖 23 种语言。没有单一模型能主导所有基准，低资源语言上的失败集中在工具调用和控制流错误，而非答案质量。低资源语言下 Agent 消耗的 token 最多约为 2 倍，且交互轮次并未增加；结构化输出任务中的语言一致性下降，切换几乎都指向英语。
In the Bible, humans tried to build the Tower of Babel but failed when language split them. Today, AI Agents are trying to "build" tasks across 23 languages—and facing the exact same barrier. 🧱🗣
Qwen Team introduces BabelArena, a new open-source multilingual Agent benchmark on 🧵👇
🔗 Dataset: https://modelscope.ai/datasets/Qwen/BabelArena
📄 Paper: https://modelscope.ai/papers/2609.23490
🏆 No single model dominates all benchmarks.
⚠ Low-resource languages show distinct failure modes: high rates of tool-use and control-flow errors (not just answer quality).
📈 Agents consume up to ~2x more tokens in low-resource languages without more interaction rounds.
🗣 Language consistency drops in structured-output tasks, with switches overwhelmingly pointing to English.
来源：ModelScope · x.com