RuneBench Skill - Smithing
Games · 2026-02-26
RuneBench 30-minute Smithing skill task. Agents play RuneScape through the SDK and are scored by the best real-game Smithing XP per minute achieved in any 15-second window.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6 Astra | 6363 |
| GPT-6.1 Sol | 1375 |
| Claude Opus 5.5 | 1063 |
| Grok 4.6 | 938 |
| Claude Fable 5.1 | 937 |
| DeepSeek V4.1 Flash | 688 |
| Gemini 3.6 Flash | 625 |
| Gemini 3.8 Flash | 625 |
| Gemini 4 Argon | 612 |
| Claude Fable 5 | 487 |
| Claude Opus 5 | 425 |
| GPT-5.6 Sol | 375 |
| GPT-6 Sol | 338 |
| Gemini 3.5 Flash | 313 |
| Gemini 3.7 Flash | 313 |
| Grok 4.7 | 313 |
| Muse Spark 1.1 | 312 |
| GPT-5.6 Terra | 281 |
| GPT-6 Luna | 250 |
| Claude Sonnet 5.5 | 225 |
| GPT-5.5 | 188 |
| Claude Haiku 5.5 | 175 |
| Muse Wonder | 162 |
| Muse Spark 1.2 | 141 |
| GLM-5.3 | 125 |
| GPT-5.4 | 119 |
| GPT-5.6 Luna | 113 |
| Opus 4.8 | 106 |
| Opus 4.7 | 81.0 |
| Kimi K2.7 Code | 81.0 |
| Sonnet 5 | 75.0 |
| Grok 4.5 | 75.0 |
| Muse Spark 1.3 | 75.0 |
| Mistral Large 4 Preview | 69.0 |
| Opus 4.6 | 63.0 |
| Gemini 3 Pro Preview | 62.0 |
| GLM-5.2 | 62.0 |
| GPT-5.4 Mini | 62.0 |
| Opus 4.5 | 56.0 |
| Kimi K3 | 50.0 |