Skip to content

Benchmark

Global PIQA

Scores are percentages. Higher is better.

Models measured
15
Scores recorded
21
Sources
2
IQ weight
2.11

Score distribution

86.8%
Mean
89.8%
Median
8.2%
Std dev
78.9%
10th percentile
92.8%
90th percentile
59.4%
Min
59.4%93.4%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.11 for chat models, measured over 15 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 3 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93.4%
2Gemini 3 Pro PreviewGemini 3 Pro Model Card · results as of November 2025 · 100 languages and cultures· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.93.4%
3Gemini 3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.8%
4Gemini 3 Flash100 languages and cultures· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.92.8%
5Qwen3.7 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.4%
6Qwen3.7 MaxQwen3.7-Max release · recommended reasoning-effort=xhigh system prompt· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.91.4%
7Qwen3.7 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.3%
8Qwen3.7 PlusQwen3.7-Plus release · recommended reasoning-effort=xhigh system prompt· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.90.3%
9Qwen3.5 397B A17B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.8%
10Qwen3.5 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.89.8%
11Qwen3.6 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.8%
12Qwen3.6 PlusQwen official release table · thinking mode unless benchmark requires otherwise· Reported by Alibaba Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.89.8%
13Seed 2.0 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.2%
14Qwen3.5 122B-A10B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.4%
15Qwen3.5 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim87.5%
16Qwen3.5 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim86.6%
17Qwen3 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim86%
18Qwen3.5 9B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim83.2%
19Qwen3.5 4B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.9%
20Qwen3.5 2B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.3%
21Qwen3.5 0.8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim59.4%

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. · Data by LLM Stats