Single-Agent Scaling Fails Multi-Agent Intelligence: Towards Foundation Models with Native Multi-Agent Intelligence

TL;DR

This paper demonstrates that scaling foundation models for single-agent tasks does not automatically produce effective multi-agent intelligence, highlighting the need for specialized development in this area.

Contribution

The paper provides empirical evidence that current large language models lack robust multi-agent capabilities and outlines future research directions for developing native multi-agent foundation models.

Findings

Scaling models improves single-agent performance but not multi-agent abilities.

Extensive experiments across 41 models and 7 benchmarks show the gap in multi-agent intelligence.

Identifies key research areas for advancing multi-agent foundation models.

Abstract

Foundation models (FMs) are increasingly assuming the role of the ''brain'' of AI agents. While recent efforts have begun to equip FMs with native single-agent abilities -- such as GUI interaction or integrated tool use -- we argue that the next frontier is endowing FMs with native multi-agent intelligence. We identify four core capabilities of FMs in multi-agent contexts: understanding, planning, efficient communication, and adaptation. Contrary to assumptions about the spontaneous emergence of such abilities, we provide extensive empirical evidence, across 41 large language models and 7 challenging benchmarks, showing that scaling single-agent performance alone does not automatically yield robust multi-agent intelligence. To address this gap, we outline key research directions -- spanning dataset construction, evaluation, training paradigms, and safety considerations -- for building FMs with native multi-agent intelligence.