Atlas

Benchmarks

← All benchmarks

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)

ModelScore
GPT-6 Astra6363
GPT-6.1 Sol1375
Claude Opus 5.51063
Grok 4.6938
Claude Fable 5.1937
DeepSeek V4.1 Flash688
Gemini 3.6 Flash625
Gemini 3.8 Flash625
Gemini 4 Argon612
Claude Fable 5487
Claude Opus 5425
GPT-5.6 Sol375
GPT-6 Sol338
Gemini 3.5 Flash313
Gemini 3.7 Flash313
Grok 4.7313
Muse Spark 1.1312
GPT-5.6 Terra281
GPT-6 Luna250
Claude Sonnet 5.5225
GPT-5.5188
Claude Haiku 5.5175
Muse Wonder162
Muse Spark 1.2141
GLM-5.3125
GPT-5.4119
GPT-5.6 Luna113
Opus 4.8106
Opus 4.781.0
Kimi K2.7 Code81.0
Sonnet 575.0
Grok 4.575.0
Muse Spark 1.375.0
Mistral Large 4 Preview69.0
Opus 4.663.0
Gemini 3 Pro Preview62.0
GLM-5.262.0
GPT-5.4 Mini62.0
Opus 4.556.0
Kimi K350.0
Loading Atlas data…