阿里巴巴发布 Qwen Intelligence,带来三款手机智能体
阿里把手机智能体做成了一套产品,三个智能体各管一摊,还把测试基准全开源了,做手机 Agent 的可以看看榜单。
阿里巴巴推出 Qwen Intelligence 平台,首发三款智能体:Mobile Planner Agent 在 MobilePA-Bench 及其 Business、Memory 三个榜单排名第一;Mobile-Use Agent 采用 API 优先、GUI 兜底的方式执行任务,MobileWorld 得分 82.1,MobileWorld-Real 得分 92.2,AndroidDaily 得分 97.2,端到端成功率达 90%;Mobile Creative Agent 可将一句话生成成品,出图耗时 3 秒,速度约为主要竞品的 2 倍。同时开放 MobilePA-Bench、MobileWorld、MobileWorld-Real、MobileWorld-Safety 四套基准测试,覆盖任务规划、跨应用执行、真机表现与安全性。
Introducing Qwen Intelligence, bringing personal intelligence within everyone's reach. 📱✨
It launches with three SOTA agents: 🥳 - Mobile Planner Agent: plans, decomposes & orchestrates complex tasks. #1 on MobilePA-Bench, MobilePA-Bench Business & Memory. - Mobile-Use Agent: gets things done, API-first with GUI fallback. MobileWorld 82.1, MobileWorld-Real 92.2, AndroidDaily 97.2, 90% end-to-end success rate. - Mobile Creative Agent: turns one sentence into ready-to-use creations. Image generated in 3s, about 2x faster than leading peers.
We're also opening up our benchmark suite: MobilePA-Bench, MobileWorld, MobileWorld-Real, and MobileWorld-Safety, covering planning, cross-app execution, real-device performance and safety.
🔗 Learn more about the agents: - Qwen Intelligence official website: https://t.co/bYHrZ18OZI - Mobile Planner Agent: https://t.co/ChQEx5fZe8 - Mobile-Use Agent: https://t.co/3vXXinCsRy - Mobile Creative Agent: https://t.co/pnnZ3uykdt
🔗 Explore our open benchmark suite: - MobilePA-Bench: https://t.co/qpEQZ6JgyE - MobileWorld (GitHub): https://t.co/h4h7da8Xl8 - Leaderboard: https://t.co/YwTn5kJfRf
🔗 More Models API: - Model Studio: https://t.co/gH6W0D8tYw