Atlas

Benchmarks

← All benchmarks

RuneBench Skill - Firemaking

Games · 2026-02-26

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

Top models (higher is better)

ModelScore
GPT-6 Astra5924
GPT-6.1 Sol4557
Gemini 4 Argon3949
Muse Spark 1.13797
GPT-6 Sol3645
Claude Opus 5.52734
DeepSeek V4.1 Flash2734
Gemini 3.8 Flash2734
Claude Opus 52632
Grok 4.52632
Claude Fable 5.12531
Gemini 3.5 Flash2227
Claude Fable 52025
Gemini 3.6 Flash1823
Grok 4.71823
Gemini 3.7 Flash1755
Muse Spark 1.21721
GLM-5.31620
GPT-6 Luna1620
Grok 4.61418
GPT-5.6 Sol1417
GPT-5.6 Terra1316
DeepSeek-V4-Pro1125
Claude Sonnet 5.51080
GLM-5.2855
GPT-5.5810
Opus 4.7600
DeepSeek-V4-Flash-0731506
GPT-5.6 Luna425
Opus 4.8370
GPT-5.3-Codex360
Muse Spark 1.3320
Muse Wonder320
GLM-5.3 Flash315
Gemini 3.5 Flash-Lite300
Kimi K2.7 Code300
Opus 4.5280
Opus 4.6280
Sonnet 4.6280
Sonnet 4.5260
Loading Atlas data…