Skip to content

Benchmark

GPQA

Science benchmarks measure how well models answer science questions. Scores are percentages. Higher is better.

Models and systems measured
364
Scores recorded
378
Sources
3
IQ weight
0.75

Score distribution

45.3%
Mean
47.5%
Median
34.4%
Std dev
2.9%
10th percentile
89%
90th percentile
0%
Min
0%97%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.75 for chat models, measured over 359 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-6.2 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim97%
2GPT-6 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim96%
3GPT-6.2 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim96%
4GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.6%
5Gemini 3.1 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.3%
6Claude Opus 4.7· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.2%
7Kimi K3.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim94.2%
8Claude Opus 4.8· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93.6%
9GPT 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93.6%
10Kimi K3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93.5%
11GPT-5.2 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93.2%
12Grok 4.5· LLM Stats data93%
13Inkling 2 Small· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93%
14GPT-5.6 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.9%
15GPT 5.4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.8%
16Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.6%
17GPT 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.4%
18Qwen3.7 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.4%
19GPT-5.6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.3%
20Hy4 preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.3%
21Gemini 3 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.9%
22Qwen 3.8 Flash Next· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.7%
23Qwen3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.7%
24Claude Opus 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.3%
25GLM 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.2%
26DeepSeek V4.1 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.9%
27Beam 1.0 501B A23B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.6%
28Kimi K2.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.5%
29Gemini 3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.4%
30Hy3 (Deep Infra)· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.4%
31Particle 1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.4%
32Qwen3.6 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.4%
33Qwen3.7 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.3%
34Claude Sonnet 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.9%
35NVIDIA Nemotron 3.5 Super VL 120B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.5%
36Inkling Small· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.5%
37Muse Spark· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.5%
38Mercury 3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.2%
39Qwen 3.8 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.2%
40Muse Spark Open· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89%
41Solar Pro 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89%
42Doubao Seed 2.0 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.9%
43Qwen3.5 397B A17B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.4%
44Qwen3.5 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.88.4%
45GPT 5.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.1%
46GPT 5.1high· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.1%
47GPT-5.1-Instant· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.1%
48GPT-5.4 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88%
49Qwen3.6 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim87.8%
50Kimi K2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim87.6%

Showing 1–50 of 374 results.

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
Qwen3.5-397B-A17B-BaseQwen3.5 release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.54.6%
Qwen3.7-Plus-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.51.5%
Qwen3.8-27B-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45%
Qwen3.8-Flash-Next-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.51.4%

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.