Atlas

Benchmarks

← All benchmarks

RuneBench Skill - Cooking

Games · 2026-02-26

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

Top models (higher is better)

ModelScore
GPT-6.1 Sol3675
Claude Opus 5.52520
GPT-6 Astra2520
GPT-6 Sol2520
Gemini 4 Argon2040
Gemini 3.7 Flash1860
GPT-5.6 Sol1470
Claude Fable 51170
Grok 4.61170
Grok 4.71110
GPT-5.51080
GLM-5.21030
Muse Spark 1.11025
GPT-6 Luna1005
Grok 4.5905
Claude Fable 5.1540
GPT-5.6 Terra540
Gemini 3 Flash Preview500
Claude Sonnet 5.5420
GLM-5.3395
Claude Haiku 5.5390
Claude Opus 5390
Gemini 3.8 Flash390
GPT-5.6 Luna375
Qwen3.7-Max (source unspecified)375
Gemini 3.5 Flash360
Gemini 3.1 Pro Preview345
Muse Spark 1.2345
Muse Wonder345
Opus 4.8270
GPT-5.3-Codex240
Muse Spark 1.3240
DeepSeek V4.1 Flash225
Kimi K3225
Sonnet 5210
Gemini 3.6 Flash210
GPT-5.4210
Opus 4.7195
Mistral Large 4 Preview120
GPT-5.4 Mini90.0
Loading Atlas data…