Minecraft AI Benchmark

An autonomous agent benchmark for defeating the Ender Dragon in Minecraft.

Description

A systematic testing harness that evaluates the performance of AI agents in a Minecraft environment, challenging them to defeat the Ender Dragon from a fresh world. The benchmark compares agents powered by Grok 4.7, Claude Opus 5.5, and GPT-6 Astra, tracking metrics such as in-game time, item collection progress (Blaze rods, Eyes of Ender), and survival rates.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.

Details

Why this is interesting: A rigorous comparison of frontier AI models acting as autonomous agents in a complex sandbox environment.
Attributed to: Raphi-2Code@R2Cdev_
Built with: Opus 5.5 Grok 4.7 GPT-6 Astra Redwood AI GPT-5.6 Sol
Tags: #benchmark #minecraft #agents #automation

Discovered: X
First posted: X · Oct 12, 2026
Listed: Added 2h ago
Build: Strong
Models: Strong