Google DeepMind says four Gemini 4 Argon agents beat Minecraft from scratch in 27 hours by writing 2,783 programs and working together
Four Gemini 4 Argon agents beat Minecraft from scratch in 27 hours. Operating in a sandbox using only code and text, they wrote 2,783 programs, divided up jobs over chat, shared notes, and debugged errors together - all with no human help ↓
Context
The test shows several AI agents splitting up a long, open ended task without human help, a step toward agents that can coordinate on real projects.
This is a first hand post from Google DeepMind quoting researcher Maxence Faldor, who shared the work first. The results come from Google's own experiment and have not been independently reproduced.
Deep dive
On October 16, Google DeepMind posted that four Gemini 4 Argon agents beat Minecraft from scratch in 27 hours. It said the agents worked in a sandbox using only code and text, wrote 2,783 programs, divided up jobs over chat, shared notes and debugged errors together with no human help.
The post quoted Maxence Faldor, who wrote earlier that day: Four Gemini 4 Argon agents, one shared Minecraft world, one goal: defeat the Ender Dragon. He said the dragon fell after 27 hours and 2,783 programs the agents wrote themselves, and linked an article describing the experiment.
Beating Minecraft means gathering resources, crafting tools, finding and entering the Nether and the End, and defeating the Ender Dragon, a long chain of steps that has been a common benchmark for AI agents. The detail that the agents acted only through code and text, rather than by seeing the screen, is how Google describes the setup.
Google made Gemini 4 Argon globally available on October 13. The posts do not say how much computing the run used or how often similar runs succeed.
What to expect next
Watch whether Google releases the code, logs or a paper so others can check how the agents split up the work and how often runs succeed.
Also watch whether rivals such as OpenAI, Anthropic or Meta report similar multi agent results, and whether the same approach is applied to tasks like software projects. Any cost figures for the 27 hour run would show how practical this kind of teamwork is.
