# Study Shows Multi‑Developer Agentic AI Beats Single‑Developer Systems

> Researchers found that combining agents built by several specialists can double accuracy on a general‑assistant benchmark.

Oossa · 2026-10-15 · https://oossa.com/en/study-shows-multi-developer-agentic-ai-beats-single-developer-systems

A new arXiv paper examines multi‑developer agentic systems (MDAs), where many programmers each create a small AI agent. When these agents are combined, they reach 70% accuracy on the GAIA benchmark, while the best single‑developer agents stay below 35%. The MDA outperforms a typical single‑developer system by 12 percentage points, even when the latter receives several times more development budget. The authors suggest MDAs could become a new way to scale AI by pooling diverse talent.

## The facts

- GAIA benchmark accuracy: 70% for MDAs vs <35% for individual agents
- MDAs beat median single‑developer systems by 12 percentage points

## Why it matters

Collaborative AI development could let smaller teams create higher‑performing assistants without massive budgets.

## Sources & references

1. [The Wealth of Agents: Scaling Agentic AI Across Developers](https://arxiv.org/abs/2610.20518) – arXiv cs.MA, 2026-10-15

Last updated: 2026-10-15