Atlas

Benchmarks

← All benchmarks

RuneBench Skill - Hitpoints

Games · 2026-02-26

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

Top models (higher is better)

ModelScore
GPT-6 Astra3235
GPT-6.1 Sol227
Claude Opus 5.5203
Gemini 4 Argon170
GPT-6 Sol163
Grok 4.5139
Gemini 3.8 Flash130
GPT-5.6 Sol126
Claude Sonnet 5.5124
Claude Opus 5114
GPT-5.6 Terra102
Claude Fable 5101
Gemini 3.7 Flash101
Opus 4.8100
DeepSeek-V4-Pro100
Grok 4.689.0
Kimi K2.7 Code86.0
Claude Fable 5.181.0
Gemini 3.5 Flash-Lite76.0
Muse Spark 1.372.0
DeepSeek V4.1 Flash70.0
Grok 4.770.0
Sonnet 569.0
GPT-5.569.0
Muse Spark 1.267.0
GPT-5.6 Luna64.0
Claude Haiku 5.563.0
DeepSeek-V4-Flash-073163.0
GPT-6 Luna63.0
Qwen3.7-Max (source unspecified)60.0
Mistral Large 4 Preview59.0
Gemini 3.6 Flash58.0
Kimi K2.648.0
Gemini 3.5 Flash40.0
GLM-5.340.0
Qwen3.8 27B39.0
Grok 4.338.0
Muse Spark 1.138.0
GLM-5.237.0
Opus 4.636.0
Loading Atlas data…