Microsoft Research拥有开源 Agensh，一种分散式的牵引装置，用于大型编译团队agents. Agensh多agent的拉带取消了大部分agent框架所依赖的中央管弦乐器。 每个工人都选择自己的子任务，分享发现，并将代码合并为通用的Git repo. 在"程序奔驰"(Programme Bench)上，球队将设置从1agent扩大至1,024agents.
TL；数据
- A Harness：从1到1,024 agents测试。 每个工人得到27.2万个托肯输入和12.8万个托肯输出限制。
- 运行在：Linux与Docker 编曲。 默认的单一节点(8名工人). 1,024-agent运行使用了16个节点，每节64个agents. 工人拥有64GiB记忆天花板。
- 性能：更多agents 融资在所有最困难的5个程序奔驰任务上得分，并更早到达。
- 最佳：79.59%的泛日通过率，GPT-6.1-索尔上为32agents. GPT-5.6-索尔命中55.06%,1024agents.
- 最糟糕的是：5个任务平均在128agents只达到28.78%. 从1到8 agents加分1.37分。
阿让什是什么？
Agensh是一个来自Microsoft Research的多agent牵引装置，它协调了许多LLM agents，没有铅的agent.
目标问题在于管弦乐团的天花板。 在管弦乐-工作设计中，一个领先的agent计划，分配和合并所有的工作。 研究论文引用了Claude Code agent团队，Codex子agents,Copilot车队和Kimi,Agent，集群的例子。 阿根什让工人自己来协调
阿根什合作如何循环？
每个工人同时同步运行着同样的5步循环：
- 收集背景：阅读目标、同行进展、信息和调查结果。
- 索赔一项次级任务：在索赔中提出，然后通过直接信息解决重叠问题。
- 采取行动：在当地利用各种工具，及早公布有用的调查结果。
- 核查结果：对照次级任务的接受标准检查工作。
- 合并进度：合并到共享Repo中，张贴更新，然后重复。
三个服务连接工人。 Gitea 托管共享的工作空间，有分支，PR和合并冲突。 Matter Most 携带一个团队通道，加上紧急碰撞的直接消息。 从DeLM改编的共享上下文板存储了指出：OBSERVED,FACT,FAIL，索赔和PATCH SUMMARI。
循环存在于即时而非运行时代码中。 除了身份证之外，每个工人都会得到同样的提示。 阿根什坐在单agent吊带上方。 这些实验使用了GitHub Copilot CLI. 研究论文指出薄的适配器可以连接其他的牵引装置，如Claude Code.
Agensh在Bench节目上的表现如何？
Microsoft团队测试了五个最难的节目Bench任务：FFmpeg,gromacs,pandoc,PHP-src和ctags. Agents从其观察到的行为重建每个程序。 他们离线工作，在6小时内。 引用重置范围从2,767个文件(pandoc)到26,266个文件(PHP-src).
GPT-5.6-sol在高推理能力下，平均最终测试-通过率移动如下：
- 1 agent: 19.31%
- 8 agents: 20.68%
- 32 agents: 26.52%
- 128 agents: 28.78%
这是9.47分的收益，大约49%的相对。 单胰岛素的得分为33.89%(1)、41.75%(8)、43.11%(32)、50.94%(128)和55.06%(1 024)。
速度也有所改善：泛神语上，128agents在30分钟的检查站通过30%. 32和8 agents需要60和90分钟。
README 新增是一个较新的模型。 随着GPT-6.1-溶液，泛氏剂从56.19%(1agent)上升到79.59%(32agents). 这是41.64%的相对收益。 这些GPT-6-Sol数字不在v1文件中。
随着团队的发展，出现了什么行为？
研究小组追踪每个尺度的工人轨迹：
- 8 agents：同行商定接口，并自行解决重叠的声称。
- 32 agents：同行审查共享工作；一次批准在反例后撤回。
- 128 agents：工人重新使用可信赖的审查员，并使一体化协议标准化。
- 1 024 agents：几名工人担任整顿员，同龄人恢复了失败的工作。
这些角色都没有硬编码。
阿根什如何与其他多agent的拉带相比？
你如何管理阿根什？
发布内容包括一个单节点的Docker编译团队和一个单agent基线。 两者都使用原产地的Copilot非加太会议。 需要Linux与Docker引擎，Compose v2，系统化，cgroup v2和Python 3.11+. 默认为8名工人，每任务6小时，gpt-5.6-Sol高功率。 您还需要程序奔驰( v1.2.5 已被验证) 和 API 模型令牌。 完整的台阶坐落在复制指南中。
关键外卖
- Agensh坐标最高达1,024 agents，没有中央管弦乐器。
- 工人通过Gitea、Mattermost和共同调查结论委员会进行自认任务。
- 128 agents将5任务程序Bench从19.31%提高到28.78%。
- GPT-6.1-溶液，32agents在胰岛素上点击79.59%.
- 代码是麻省理工学院的许可，但象征性成本不披露。
请检查access-date=中的日期值 (帮助) 页面存档备份，存于互联网档案馆页面存档备份，存于互联网档案馆 GitHub Repo and Project Page. 并加入我们的通讯。 等等你在电报里吗现在你也可以加入我们的电报。
需要和我们合作推广您的 GitHub Repo OR Hugging Face Page OR Product Release OR Webinar 等吗？联系我们
Asif Razzaq是Marktechpost AI Media Inc.的首席执行官。 作为一个有远见的企业家和工程师，阿西夫致力于利用人工智能的潜力促进社会福祉。 他最近的努力是人工智能媒体平台的发布，该平台以深入报道机器学习和深层学习新闻而著称，这些新闻在技术上既可靠又为广大观众所容易理解。 该平台每月有超过200万次观看，表明其受观众欢迎。
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Microsoft Research has open-sourced Agensh, a decentralized harness for large teams of coding agents. The Agensh multi-agent harness removes the central orchestrator that most agent frameworks depend on. Every worker picks its own sub-task, shares findings and merges code into a common Git repo. On ProgramBench, the team scaled the setup from 1 agent to 1,024 agents.
TL;DR
- A Harness: Tested from 1 to 1,024 agents. Each worker gets a 272,000-token input and 128,000-token output limit.
- Runs on: Linux with Docker Compose. Single node by default (8 workers). The 1,024-agent run used 16 nodes, 64 agents each. Workers have a 64 GiB memory ceiling.
- Performance: more agents raised scores on all 5 hardest ProgramBench tasks, and reached them sooner.
- Best: 79.59% pandoc pass rate with 32 agents on GPT-6.1-sol. GPT-5.6-sol hit 55.06% with 1,024 agents.
- Worst: the 5-task mean reached only 28.78% at 128 agents. Going from 1 to 8 agents added 1.37 points.
What is Agensh?
Agensh is a multi-agent harness from Microsoft Research that coordinates many LLM agents without a lead agent.
The target problem is the orchestrator ceiling. In orchestrator-worker designs, a lead agent plans, assigns and merges all work. The research paper cites Claude Code agent teams, Codex sub-agents, Copilot fleet and Kimi Agent Swarm as examples. Agensh lets the workers do that coordination themselves.
How does the Agensh cooperation loop work?
Each worker runs the same 5-step loop, concurrently and asynchronously:
- Gather context: read the goal, peer progress, messages and findings.
- Claim a sub-task: post a CLAIM, then settle overlaps by direct message.
- Take action: work locally with tools and post useful findings early.
- Verify results: check the work against the sub-task’s acceptance criteria.
- Merge progress: merge into the shared repo, post an update, then repeat.
Three services connect the workers. Gitea hosts the shared workspace, with branches, PRs and merge conflicts. Mattermost carries a team channel plus direct messages for urgent collisions. A shared context board, adapted from DeLM, stores typed notes: OBSERVED, FACT, FAIL, CLAIM and PATCH_SUMMARY.
The loop lives in the prompt, not in runtime code. Every worker gets the same prompt except its ID. Agensh sits on top of a single-agent harness. The experiments used GitHub Copilot CLI. The research paper states thin adapters can connect other harnesses, such as Claude Code.
How well does Agensh scale on ProgramBench?
Microsoft team tested the 5 hardest ProgramBench tasks: FFmpeg, gromacs, pandoc, PHP-src and ctags. Agents rebuild each program from its observed behavior. They work offline, within 6 hours. Reference repos range from 2,767 files (pandoc) to 26,266 files (PHP-src).
With GPT-5.6-sol at high reasoning effort, the mean final test-pass rate moved as follows:
- 1 agent: 19.31%
- 8 agents: 20.68%
- 32 agents: 26.52%
- 128 agents: 28.78%
That is a 9.47-point gain, about 49% relative. On pandoc alone, scores were 33.89% (1), 41.75% (8), 43.11% (32), 50.94% (128) and 55.06% (1,024).
Speed improved too: on pandoc, 128 agents passed 30% at the 30-minute checkpoint. 32 and 8 agents needed 60 and 90 minutes.
The README adds a newer model. With GPT-6.1-sol, pandoc rose from 56.19% (1 agent) to 79.59% (32 agents). That is a 41.64% relative gain. These GPT-6.1-sol figures are not in the v1 paper.
What behaviors emerged as the team grew?
The research team traced worker trajectories at each scale:
- 8 agents: peers agreed on interfaces and resolved overlapping claims themselves.
- 32 agents: peers reviewed shared work; one approval was withdrawn after a counterexample.
- 128 agents: workers reused trusted reviewers and standardized an integration protocol.
- 1,024 agents: several workers acted as integrators, and peers recovered failed work.
None of these roles was hard-coded.
How does Agensh compare with other multi-agent harnesses?
How do you run Agensh?
The release includes a single-node Docker Compose team and a single-agent baseline. Both use native Copilot ACP sessions. You need Linux with Docker Engine, Compose v2, systemd, cgroup v2 and Python 3.11+. Defaults are 8 workers, 6 hours per task and gpt-5.6-sol at high effort. You also need ProgramBench (v1.2.5 was validated) and a model API token. Full steps sit in the reproduction guide.
Key Takeaways
- Agensh coordinates up to 1,024 agents with no central orchestrator.
- Workers self-claim tasks via Gitea, Mattermost and a shared findings board.
- 128 agents lifted the 5-task ProgramBench mean from 19.31% to 28.78%.
- GPT-6.1-sol with 32 agents hit 79.59% on pandoc.
- Code is MIT licensed, but token cost is not disclosed.
Check out the Paper, GitHub Repo and Project Page. Also, feel free to follow us on Twitter and don’t forget to join our 150k+ML SubReddit and Subscribe to our Newsletter. Wait! are you on telegram? now you can join us on telegram as well.
Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us
Asif Razzaq is the CEO of Marktechpost AI Media Inc.. As a visionary entrepreneur and engineer, Asif is committed to harnessing the potential of Artificial Intelligence for social good. His most recent endeavor is the launch of an Artificial Intelligence Media Platform, Marktechpost, which stands out for its in-depth coverage of machine learning and deep learning news that is both technically sound and easily understandable by a wide audience. The platform boasts of over 2 million monthly views, illustrating its popularity among audiences.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.