Meta 论文提出 GitSwarm：多智能体共享 Git 仓库协作
New Meta paper shows that AI agents can build on each other's work, even failed attempts, when every step lives in a shared Git repo.
Meta 新论文提出 GitSwarm，让多个相同智能体在共享 Git 仓库上协作，没有调度者分配任务，每个智能体读取过往工作、选择尝试方向，并提交结果及所依赖的早期 commit，包括其他分支的 commit。在 ProgramBench 的 50 个程序重建任务上，GitSwarm 得分 79.4%，而单个 Codex 智能体在相近算力下持续工作最高为 65.1%；后续智能体建立在 94.7% 的已保存贡献之上。论文指出，多数给智能体增加算力的方式把每次运行视为独立，运行结束后部分结果和失败都会消失。
正文
当前显示原文，中文全文译文尚未就绪。
New Meta paper shows that AI agents can build on each other's work, even failed attempts, when every step lives in a shared Git repo.
Most ways to give agents more compute treat each run on its own. When a run ends, its partial results and failures vanish, so later runs may need to rediscover them.
GitSwarm runs many identical agents on a shared repo, with no boss assigning tasks. Each agent reads past work, picks what to try, and commits its result with a list of earlier commits it used, even from other branches.
On 50 program-rebuilding tasks from ProgramBench, GitSwarm scored 79.4%, while a single Codex agent told to keep working peaked at 65.1% at similar compute. Later agents built on 94.7% of saved contributions.
Instead of pushing a single agent to keep going, run several over a shared Git history that keeps every attempt, failures included.