# AgentWorld: Benchmarking Long-Horizon Collaboration of Multi-agent LLMs

> Source: <https://aiflash.com/news/127584/>
> Published: 2026-09-28 03:30:02+00:00

Existing multi-agent benchmarks primarily test in competitive settings, short-horizon interactions under 20 steps, or simply aggregate individual performance, failing to isolate and highlight genuine collaboration capabilities of LLM-based agents. We introduce AgentWorld, a benchmark of 100 human-an