Team of 8 AI agents beats best solo agent building Colosseum in Minecraft (0.72 vs 0.58)
DimitrisPapail · x · 2026-10-02
A developer ran an experiment building the Colosseum in Minecraft with 16 AI agents: 8 communicated via a shared messaging board as a team, while 8 worked solo in parallel. The team's worst agent (0.72) beat the best solo performer (0.58), all running on Qwen 3.8 27B via RTX 5090s.
It's a striking demo of test-time communication paying off — agents sharing information mid-task acts as a capability amplifier, which the author frames as the beginning of an era of 'Swarm engineering'.
More from coding & agent
- Dev discovers his game-testing agent adds two humans to check mixed-reality accessibility for kids and adults — nptacek · 2026-10-02
- Developer quip: the more Codex agents I run, the more work I end up doing — LauraModiano · 2026-10-02
- Grok Build Adds Agent Dashboard to Manage Multiple Parallel Coding Agents From One Screen — XFreeze · 2026-10-02
- Andriy Burkov slams Supabase: 150s edge function cap, no npm, no CLI logs — burkov · 2026-10-02
- Dev argues agent-written Playwright tests just bloat codebases; verification engineering is the real path — KlausCodes · 2026-10-02
- Why AI hasn't disrupted office work: desktop apps lack APIs agents can use — mymooh · 2026-10-02