To match a 100x single-agent token scale-up, a swarm needs 900x-15,000x more agents
tobyordoxford · x · 2026-09-22
Toby Ord quantifies why swarms rarely pay off: a 4-agent swarm uses 2x total tokens but only 0.5x per agent, so it can run up to 2x faster (never quite reaching that in practice). The catch compounds at scale — matching the performance gain of a 100x token scale-up for a single agent requires growing the swarm by 900x to 15,000x, and the gap widens further with larger scale-ups. Swarms buy speed, not performance.
More from coding & agent
- Open-source landing-page-design skill packs intake, copy and visual system for Claude Code — GCWebDesigner · 2026-09-22
- Doe v. GitHub AI case sees a meaningful ruling on Copilot output claims — technollama · 2026-09-22
- claude-uds-bridge: An MCP Server Connecting Codex and Claude Natively — LeonKohli · 2026-09-22
- Generating responsive UIs in under 2 seconds with shadcn plus Mobbin MCP — msharmas · 2026-09-22
- Nat Friedman admits Meta's Muse was inspired by OpenClaw, bought hundreds of Mac minis for the team — EdwardSun0909 · 2026-09-22
- SkillLift cuts agent skill-evolution token cost 40-70% by ranking, not rollouts — dair_ai · 2026-09-22