Skip to content

Benchmark

RNG-Bench · Matching Pairs Responses per Score

Lower scores are better.

Models and systems measured
5
Scores recorded
5
Sources
1
IQ weight
No weight published

Score distribution

12.5
Mean
11.6
Median
4
Std dev
8
10th percentile
19.7
90th percentile
8
Min
819.7
The band covers the middle 80% of scores. The line marks the median.

Results

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

Model or systemScore
Gemini 3.1 Pro Previewsingle-player · 10×10 · image · noise theme· Reported by RNG-Bench / InternLM· via BenchBoard docs10
GPT 5.4single-player · 10×10 · image · noise theme· Reported by RNG-Bench / InternLM· via BenchBoard docs8
Kimi K2.5single-player · 10×10 · image · noise theme· Reported by RNG-Bench / InternLM· via BenchBoard docs13.2
Qwen3.5 397B A17Bsingle-player · 10×10 · image · noise theme· Reported by RNG-Bench / InternLM· via BenchBoard docs19.7
Seed-2.0-Litesingle-player · 10×10 · image · noise theme· Reported by RNG-Bench / InternLM· via BenchBoard docs11.6

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.