RuneBench Skill - Fishing
Games · 2026-02-26
RuneBench 30-minute Fishing skill task. Agents play RuneScape through the SDK and are scored by the best real-game Fishing XP per minute achieved in any 15-second window.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 1125 |
| Gemini 3 Flash Preview | 1085 |
| Grok 4.5 | 1070 |
| Gemini 3.5 Flash | 835 |
| GPT-5.4 Mini | 830 |
| GPT-5.6 Sol | 810 |
| GPT-5.5 | 780 |
| GPT-5.4 | 755 |
| Opus 4.7 | 585 |
| Opus 4.5 | 445 |
| GPT-5.6 Luna | 425 |
| DeepSeek-V4-Pro | 415 |
| GLM-5.2 | 380 |
| Sonnet 4.5 | 355 |
| Sonnet 5 | 340 |
| Opus 4.8 | 320 |
| Gemini 3.1 Pro Preview | 295 |
| Qwen3.7-Max | 290 |
| GPT-5.3-Codex | 280 |
| Kimi K2.7 Code | 260 |
| Gemini 3 Pro Preview | 255 |
| Grok 4.3 | 165 |
| Opus 4.6 | 110 |
| Sonnet 4.6 | 80.0 |
| Kimi K2.6 | 75.0 |
| Qwen3 Max (rolling alias) | 20.0 |
| Haiku 4.5 | 0.0 |
| GLM-5 | 0.0 |
| GPT-5.4 Nano | 0.0 |
| Kimi K2.5 | 0.0 |
| Qwen3.5 35B A3B | 0.0 |