Bespoke Labs benchmark finds multi-agent AI teams barely beat single agents
Bespoke Labs' MAC Bench, a 12-task benchmark, separates gains from adding agents from gains from coordination, finding coordination adds almost nothing for frontier models.
Read the full story at AlphaSignal ↗
Timeline · 1 report
- 2026-11-18 17:00 · AlphaSignal
Bespoke Labs' MAC Bench Reveals 8-Agent AI Teams Barely Outperform Solo Workers