Skip to content

Benchmark

RecreationBench paper · Overall · Applications with Prog ≥90%

Higher scores are better.

Models and systems measured
10
Scores recorded
10
Sources
1
IQ weight
1.65

Score distribution

4
Mean
2
Median
4.8
Std dev
1.6
10th percentile
5.5
90th percentile
0
Min
017.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.65 for chat models, measured over 8 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
1GPT-6 AstraRecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · max reasoning effort per site data · 272k context · 128k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs17.6
2Claude Opus 5RecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · max reasoning effort per site data · 1m context · 128k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs5.5
3GPT-5.6 SolRecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · max reasoning effort per site data · 272k context · 128k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs5.2
4Gemini 3.7 FlashRecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · high reasoning effort per site data · 1m context · 64k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs2.4
5GLM 5.3RecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · max reasoning effort per site data · 1m context · 128k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs2
6Kimi K3RecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · max reasoning effort per site data · 1m context · 128k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs2
7Claude Opus 4.8RecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · max reasoning effort per site data · 1m context · 128k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs1.6
8Qwen3.7 PlusRecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · thinking enabled per site data · 256k context · 32k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.0

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
Grok 4.6RecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · xhigh reasoning effort per site data · 500k context · 128k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs1.6
Qwen3.8 Max 0902RecreationWorld benchmark-owner paper run · 250 held-out tasks · GUI exploration + coding tools + visual verification · xhigh reasoning effort per site data · 256k context · 32k max output tokens · Overall · 250 held-out tasks · equal-weight platform macro average· Reported by QwenLM / RecreationWorld authors· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.2

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.