A team of Gemini 4 Argon agents beats Minecraft
Four Gemini 4 Argon agents autonomously defeated Minecraft's Ender Dragon in 27 hours using only Mineflayer.
Four Gemini 4 Argon agents were placed in a sandboxed Minecraft survival world with the sole goal of defeating the Ender Dragon. Each agent used the Mineflayer library and coordinated through in-game chat and shared memory, planning, writing, and debugging every program without assigned roles or human-written game logic. Starting on Easy from a random world, they wrote 2,783 programs, died 605 times, and won in 27 hours and 14 minutes. The authors call it the first end-to-end autonomous Minecraft completion using only standard Mineflayer, and discuss limits of long-horizon multi-agent coordination.
- Four Gemini 4 Argon agents shared one Minecraft survival goal.
- Agents used Mineflayer, in-game chat, and shared memory.
- They wrote 2,783 programs and died 605 times.
- The Ender Dragon fell in 27 hours 14 minutes on Easy difficulty.
- Researchers could pause the run but did not intervene.
Full article209 words · extracted from arxiv.org · click to collapse
We placed four Gemini 4 Argon agents in a sandboxed Minecraft survival world and gave them one goal: defeat the Ender Dragon. Each agent controlled a player via an open-source library called Mineflayer and coordinated with its teammates through in-game chat and shared memory. No one assigned roles or wrote game logic. The agents planned, wrote, and debugged every program themselves. Researchers could watch the run on a live dashboard and pause or stop it at any time, but they had no cause to intervene. Starting from a randomly generated world on Easy difficulty, the team wrote 2,783 programs and defeated the dragon in 27 hours and 14 minutes, dying 605 times along the way. To our knowledge, this is the first end-to-end autonomous completion of Minecraft by AI agents, achieved using only the standard Mineflayer library without pre-scouted seeds or other game-specific tools written by humans. Winning required the agents to keep a coherent plan across thousands of dependent steps for more than a day. Complex real-world software engineering and enterprise tasks demand the same long-horizon coding and multi-agent coordination. We describe the setup, how the run unfolded, the team's strategies (dividing the work, sharing what they learned, recovering each other's items), and the limits of the result.
Text extracted automatically; images, tables and formatting may be missing. Original: https://arxiv.org/abs/2610.22319