Orak - Minecraft: leaderboard
Metric: Game Score (0-100). Source: krafton-ai.github.io. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 03-25) | 75 |
| 2 | Claude 3.7 Sonnet | 75 |
| 3 | O3 Mini (2025-01-31) | 75 |
| 4 | GPT-4o (2024-08-06) | 71 |
| 5 | GPT-4o Mini | 46 |
| 6 | DeepSeek R1 | 41.7 |
| 7 | Llama 3.2 3B Instruct | 0 |
| 8 | Qwen 2.5 7B Instruct | 0 |
| 9 | Llama 3.2 1B Instruct | 0 |
| 10 | Qwen 2.5 3B Instruct | 0 |
Interactive version: theaggregate.ai/benchmark?slug=orak-minecraft · How It Works · Data refreshed daily, snapshot 2026-09-19.