Four Gemini 4 Argon agents beat Minecraft's Ender Dragon autonomously in 27 hours
Maxence Faldor · hf · 2026-10-16
A Google DeepMind team placed four Gemini 4 Argon agents in a sandboxed Minecraft survival world with one goal: defeat the Ender Dragon.
- Each agent controlled a player via the open-source Mineflayer library and coordinated through in-game chat and shared memory; no roles were assigned and no game logic was written by humans — agents planned, wrote, and debugged every program themselves
- Starting from a randomly generated world on Easy difficulty, the team wrote 2,783 programs and defeated the dragon in 27 hours 14 minutes, dying 605 times along the way
- To the authors' knowledge, this is the first end-to-end autonomous completion of Minecraft by AI agents, using only standard Mineflayer — no pre-scouted seeds or human-written game-specific tools
- Winning required keeping a coherent plan across thousands of dependent steps for over a day — the same long-horizon coding and multi-agent coordination demanded by real-world software engineering. The write-up details the setup, the run, team strategies (task division, knowledge sharing, item recovery), and limitations
More from coding & agent
- SpecAgent: small model drafts tool calls, large model verifies in parallel to cut latency — Zilin Wang · 2026-10-16
- Salesforce's CLIFT turns sparse judge feedback into rewards for open-source web agents — Salesforce · 2026-10-16
- Agent spots invalid SPF record and autonomously fixes it in Vercel DNS via browser — dkundel · 2026-10-16
- MCPJam engineer asks how to test if ChatGPT plugins actually see latest UI selection — Paoli99 · 2026-10-16
- ChatGPT revives unplayable 2000s game Bookworm on Mac and iPhone in 15 minutes — Shleemy_Pants · 2026-10-16
- Agentflow: open-source collective memory network for agents via FastMCP and pgvector — agentflow0328 · 2026-10-16