Agensh: Scaling Organizational Intelligence to 1,024 Agents
Agensh scales self-organized multi-agent work to 1,024 agents without a central orchestrator.
Agensh is a multi-agent harness in which concurrent workers gather context, claim sub-tasks, act, verify, and merge results without a central orchestrator. Infrastructure includes a shared workspace, a message interface, and shared context. Using GPT-5.6-sol, scaling from 1 to 128 agents raised the mean pass rate on the five hardest ProgramBench tasks from 19.31% to 28.78%. On pandoc, scaling from 1 to 1,024 agents raised the rate from 33.89% to 55.06%.
- Workers self-assign tasks without a central orchestrator.
- Shared workspace, messaging, and context support the loop.
- 128 agents lift ProgramBench pass rate from 19.31% to 28.78%.
- 1,024 agents lift pandoc pass rate from 33.89% to 55.06%.
- Products
- Agensh
- AI models
- GPT-5.6-sol
From the original article
1 min read at huggingface.co
A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness frameworks support multi-agent systems. However, the scalability of current multi-agent harnesses is often constrained by a central orchestrator's capacity to allocate tasks and coordinate workers.
By Zhihao Zhan, Ting Song, Li Dong · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.