Atlas

Benchmarks

← All benchmarks

RuneBench Skill - Ranged

Games · 2026-02-26

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

Top models (higher is better)

ModelScore
GPT-6 Astra748
Gemini 4 Argon600
Claude Opus 5.5510
GPT-6.1 Sol492
GPT-6 Sol398
Claude Opus 5376
Claude Fable 5.1358
Gemini 3.8 Flash316
Grok 4.6316
Gemini 3.7 Flash308
Claude Fable 5248
Sonnet 5246
GPT-5.6 Terra246
Gemini 3.5 Flash240
Opus 4.7228
DeepSeek V4.1 Flash204
Grok 4.5204
Claude Sonnet 5.5200
GLM-5.3182
DeepSeek-V4-Flash-0731166
Muse Spark 1.2146
GPT-5.6 Sol144
Gemini 3 Flash Preview138
GPT-5.5134
Gemini 3.6 Flash130
Opus 4.8128
GPT-5.6 Luna122
Gemini 3.1 Pro Preview118
GLM-5.2116
Opus 4.6114
GPT-5.3-Codex84.0
Muse Wonder84.0
Grok 4.782.0
Muse Spark 1.180.0
Claude Haiku 5.564.0
GPT-6 Luna64.0
Grok 4.364.0
Opus 4.562.0
Gemini 3.5 Flash-Lite50.0
Sonnet 4.542.0
Loading Atlas data…