OpenMobile-2 证明开源智能手机 agents可以对抗专有模式
关键点
- 开源 OpenMobile-2项目在超过110个模拟应用的环境下对AI智能手机agents进行列车。
- 应用多样性比数据量更重要。 有了同样数量的培训数据，应用程序数量的增加将未见的成功率从60.7%推到67.5%。
- 在真正的Android设备上，较大的型号在SPA-Bench上得分59.6分，是开放型号中最高的。
仅靠更多的数据并不能使智能手机agents更好。 以同样数量的数据，最重要的是agent列车上有多少不同的应用。
AI agents操作的应用软件主要通过实践学习，但他们需要的数据大多被锁定在大公司。 开放模式因此落后，撰写了南京大学，AI公司SenseTime等多所大学的研究人员。
OpenMobile-2是他们试图缩小这一差距的尝试。 他们释放了两款以Qwen3.5为主建的agent型号，配有90亿和Qwen3.6等参数，同时释放了完整的训练环境，数据，以及复制一切的指示。 作者说，他们的模型排名在目前智能手机agents中接近顶级。
为何110个应用软件的重建胜过真正的应用
有关真实应用的培训很痛苦，因为登录，真实购买，而且无法重设状态，使得这一过程昂贵而难以重复。 于是研究人员重建了主要在中国流行的35个应用软件，包括道宝，JD,DiDi和Meituan等，加上全球使用的Slack和Google Drive等服务。 结合Android模拟器中现有的模拟和应用程序运行，练习地覆盖了110多个应用。
人文评论员将被重建的应用与原作相提并论，并不断精炼，直到工作流程和界面相匹配。 命令、预算和旅行计划贯穿各会，数据可在应用程序之间通过剪贴板传递。 每一次模拟都会自动重回其起步状态，而agent只看到截图，无法访问内部应用数据，模仿一个人如何实际使用。
从这个设置中，团队产生了大约11,600个实例工作流程和超过2,000个练习任务，并进行自动化的成功检查。 Google's Gemini 3.1 (英语). Pro被作为产生实例的教师模型，这意味着OpenMobile-2正在有效地从一个封闭的系统中学习。
真电话的性能跳得最多
这两种模式都显示在既定的基准测试上有明显的收益。 在AndroidWorld上，更大的模型从67.1到79.9分跳出，而较小的则从54.3到74分。 前身OpenMobile-1在同一测试中打出64.7分。 在更硬的MobileWorld 基准测试上，它覆盖了跨越多个应用的更长的任务，更大的模型从35.9分提高到50.4分。
在真实的Android设备上最大的跳出显示，更大的型号从SPA-Bench上从31.9分攀升到59.6分。 研究者说，这是开放模型中最好的分数。
培训环境多样性不仅仅是这里的数据量。 研究者持有数据数量恒定，只增加了agent所应用的应用，这把未见应用的成功率从60.7%推到67.5%。 以同样的数据预算，应用品种比起少数应用程序的额外例子。
顶上强化学习只带来有限的收益，主要是在类似于培训设置的环境中。 在AndroidWorld上，更大的型号实际上在之后得分稍差，以79.9分而不是80.8分登陆。
直接命令让 agents 跳过菜单迷宫
研究的一个实验部分超越了敲打和打扫。 超过50个模拟应用提供了"添加到推车"等直接指令，因此agent可以通过菜单跳过导航，而agent则根据可用的条件在两种方法之间自由切换。
不过，单靠自己，这并没有多大帮助，因为模型往往忽略了没有定向训练的捷径。 一旦他们接受了有关他们的训练，agents就很好地使用它们，解决了高达10.3%多的任务，同时需要减少大约五个步骤。 研究人员说，这种能力也转移到训练期间从未遇到的模型的工具上。
在后期的强化学习阶段，每当跑步包括至少一个直接指令时，工具使用都会得到奖金因素，尽管奖励仍然取决于任务是否实际完成。 当agent报告成功时，惩罚被踢出，这不是真的。 这些接口目前仅存在于模拟应用中，研究者称该方法本身为实验。
部分结果出自团队以215项任务所构建的定制基准测试，其中近一半跨越多个应用。 任务完成不是由一个语言模型来判断，而是由一个程序来比较应用程序每次运行前后存储的数据。 提交人还检查了培训与测试任务之间的重叠，只发现一些小的相似之处。
Microsoft 时速研究已经将API agents归类为更可靠，GUI agents在2025年的研究中更灵活，指出移动应用是不存在外部接口的典型案例。 Oppo用它开放的agent X-OmniClaw使用不同的路线，它运行在真实的设备上，并且通过深链接重放之前在Taobao或Meituan等应用中显示的路径而不再通过菜单来点击。 在封闭型号中，Google已经将智能手机控制直接建入了Gemini 3.5 Flash.
AI News Without the Hype – 由人类治愈
订阅THE DECODER免费阅读，每周AI通讯，我们独家的"AI雷达"前沿报告每年6次，全面存档访问，访问我们的评论部分。
现在订阅
本文由 AI宝库自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
OpenMobile-2 proves open-source smartphone agents can rival proprietary models
Key Points
- The open-source OpenMobile-2 project trains AI smartphone agents in an environment with more than 110 simulated apps.
- App diversity matters more than data volume. With the same amount of training data, increasing the number of apps pushed the success rate on unseen applications from 60.7 to 67.5 percent.
- On real Android devices, the larger model scored 59.6 points on SPA-Bench, the highest result among open models.
More data alone doesn't make smartphone agents better. With the same amount of data, what matters most is how many different apps an agent trains on.
AI agents that operate apps on their own learn mainly through practice, but the data they need is mostly locked up at big companies. Open models lag behind as a result, write researchers from Nanjing University, AI company SenseTime, and several other universities.
OpenMobile-2 is their attempt to close that gap. They're releasing two agent models built on Qwen3.5 with 9 billion and Qwen3.6 with 27 billion parameters, along with the full training environment, data, and instructions to reproduce everything. The authors say their model ranks near the top among current smartphone agents.
Why rebuilding 110 apps beats training on real ones
Training on real apps is painful because logins, real purchases, and no way to reset state make the process expensive and hard to repeat. So the researchers rebuilt 35 apps popular mainly in China, including Taobao, JD, DiDi, and Meituan, plus globally used services like Slack and Google Drive. Combined with existing simulations and apps running in an Android emulator, the practice ground covers more than 110 applications.
Human reviewers compared the rebuilt apps to the originals and kept refining until workflows and interfaces matched. Orders, budgets, and travel plans persist across sessions, and data can pass between apps via the clipboard. Every simulation resets to its starting state automatically, while the agent only sees screenshots and has no access to internal app data, mimicking how a person would actually use them.
From this setup, the team generated about 11,600 example workflows and more than 2,000 practice tasks with automated success checks. Google's Gemini 3.1 Pro served as the teacher model that produced the examples, which means OpenMobile-2 is effectively learning from a closed system.
Real-phone performance jumps the most
Both models show clear gains on established benchmarks. On AndroidWorld, the larger model jumps from 67.1 to 79.9 points, while the smaller one goes from 54.3 to 74. The predecessor OpenMobile-1 scored 64.7 on the same test. On the harder MobileWorld benchmark, which covers longer tasks spanning multiple apps, the larger model improves from 35.9 to 50.4 points.
The biggest jump shows up on real Android devices, where the larger model climbs from 31.9 to 59.6 points on SPA-Bench. The researchers say that's the best score among open models.
Training environment diversity matters more than sheer data volume here. The researchers held the amount of data constant and only increased the number of apps the agent practiced on, which pushed the success rate on unseen apps from 60.7 to 67.5 percent. With the same data budget, app variety beats extra examples from a handful of apps.
Reinforcement learning on top brought only limited gains, mainly in environments that resembled the training setup. On AndroidWorld, the larger model actually scored slightly worse afterward, landing at 79.9 instead of 80.8 points.
Direct commands let agents skip the menu maze
An experimental part of the study goes beyond tapping and swiping. More than 50 of the simulated apps offer direct commands like "add to cart" so the agent can skip navigating through menus, and the agent freely switches between both approaches depending on what's available.
On its own, though, this doesn't help much because the models tend to ignore the shortcuts without targeted training. Once they've been trained on them, the agents use them well, solving up to 10.3 percentage points more tasks while needing about five fewer steps. The researchers say this ability also transfers to tools the models never encountered during training.
During the later reinforcement learning phase, tool use got a bonus factor whenever a run included at least one direct command, though the reward still depended on whether the task was actually completed. Penalties kicked in when the agent reported a success that wasn't real. These interfaces only exist in the simulated apps for now, and the researchers call the approach experimental themselves.
Some results come from a custom benchmark the team built with 215 tasks, nearly half of which span multiple apps. Task completion isn't judged by a language model but by a program that compares the app's stored data before and after each run. The authors also checked for overlap between training and test tasks and found only minor similarities.
Microsoft Research had already classified API agents as more reliable and GUI agents as more flexible in a 2025 study, pointing to mobile apps as a typical case where external interfaces don't exist. Oppo takes a different route with its open agent X-OmniClaw, which runs on real devices and replays previously shown paths in apps like Taobao or Meituan via deeplinks instead of clicking through menus again. Among closed models, Google has built smartphone control directly into Gemini 3.5 Flash.
AI News Without the Hype – Curated by Humans
Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.
Subscribe now
Auto-fetched and translated by AI Gem Box. All rights belong to the original author.