Skip to content

Benchmark

Seal-0

Scores are percentages. Higher is better.

Models measured
7
Scores recorded
9
Sources
2
IQ weight
0.27

Score distribution

49.4%
Mean
46.9%
Median
5.7%
Std dev
44.1%
10th percentile
57.4%
90th percentile
41.4%
Min
41.4%57.4%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.27 for chat models, measured over 7 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Kimi K2.5setting not named by the source· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.57.4%
2Kimi K2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim57.4%
3Kimi K2 Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim56.3%
4Qwen3.5 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim47.2%
5Qwen3 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim46.9%
6Qwen3.5 397B A17B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim46.9%
7Qwen3.5 397B A17BQwen3.5 release/model card · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.46.9%
8Qwen3.5 122B-A10B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim44.1%
9Qwen3.5 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim41.4%

Data credits: Data by LLM Stats · 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.