RuneBench Skill - Mining
Games · 2026-02-26
RuneBench 30-minute Mining skill task. Agents play RuneScape through the SDK and are scored by the best real-game Mining XP per minute achieved in any 15-second window.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 650 |
| GPT-5.5 | 650 |
| GPT-5.6 Sol | 438 |
| Gemini 3.1 Pro Preview | 385 |
| Grok 4.5 | 368 |
| Gemini 3 Pro Preview | 367 |
| GPT-5.4 Mini | 367 |
| Opus 4.7 | 350 |
| Opus 4.8 | 350 |
| Sonnet 5 | 350 |
| GLM-5.2 | 333 |
| GPT-5.4 | 333 |
| Qwen3.7-Max | 332 |
| GPT-5.6 Luna | 325 |
| Opus 4.5 | 315 |
| DeepSeek-V4-Pro | 315 |
| Gemini 3.5 Flash | 298 |
| GPT-5.3-Codex | 298 |
| Kimi K2.7 Code | 280 |
| Opus 4.6 | 175 |
| Gemini 3 Flash Preview | 175 |
| Grok 4.3 | 140 |
| Kimi K2.6 | 123 |
| Haiku 4.5 | 79.0 |
| Sonnet 4.5 | 70.0 |
| Kimi K2.5 | 53.0 |
| Sonnet 4.6 | 44.0 |
| GLM-5 | 0.0 |
| GPT-5.4 Nano | 0.0 |
| Qwen3.5 35B A3B | 0.0 |
| Qwen3 Max (rolling alias) | 0.0 |