Atlas

Benchmarks

← All benchmarks

RuneBench Skill - Attack

Games · 2026-02-26

RuneBench 30-minute Attack skill task. Agents play RuneScape through the SDK and are scored by the best real-game Attack XP per minute achieved in any 15-second window.

Top models (higher is better)

ModelScore
GPT-6.1 Sol6875
GPT-6 Astra2420
Gemini 4 Argon2414
Grok 4.72414
Claude Fable 5.12412
Claude Opus 5.5446
Gemini 3.5 Flash436
GPT-6 Sol420
Claude Opus 5352
GPT-5.6 Sol350
Grok 4.5326
Gemini 3.8 Flash284
Claude Fable 5282
Gemini 3.7 Flash254
GPT-5.6 Terra246
Grok 4.6210
Sonnet 5196
Opus 4.8192
Claude Sonnet 5.5192
DeepSeek V4.1 Flash192
Gemini 3.6 Flash168
GPT-5.6 Luna126
Gemini 3.5 Flash-Lite120
DeepSeek-V4-Flash-0731114
GPT-5.5110
DeepSeek-V4-Pro108
Muse Spark 1.1106
Opus 4.7100
Gemini 3 Flash Preview96.0
Muse Spark 1.292.0
Sonnet 4.586.0
GLM-5.286.0
Muse Spark 1.386.0
Inkling84.0
Kimi K2.584.0
Qwen3.8 27B84.0
Grok 4.382.0
GLM-5.380.0
Qwen3.7-Max (source unspecified)74.0
GLM-5.3 Flash72.0
Loading Atlas data…