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
101Claude Opus 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79.6%
102Nova 2 Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79.6%
103Nemotron 3 Super 120B A12BNeMo Evaluator / mostly NeMo Skills · no tools· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.79.2%
104GLM 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79.1%
105O1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79%
106Seed 2.0 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79%
107Gemma 4 12B It· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.8%
108Sarvam 105B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.7%
109MiniMax M2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.78%
110o1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78%
111Qwen3 235B A22B Instruct 2507· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim77.5%
112o3 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim77.2%
113Qwen3 Next 80B A3Bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim77.2%
114Qwen3.5 Omni FlashQwen3.5-Omni release · no-thinking setting· Reported by Qwen, Qwen Team· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.76.4%
115Qwen3.5 4B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim76.2%
116Llama 3.1 Nemotron Ultra 253B v1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim76%
117Kimi K2 0905· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.8%
118Nemotron 3.5 Lightning 30B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.4%
119Claude Sonnet 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.4%
120GLM 4.7 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.2%
121Kimi K2 Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.1%
122GLM 4.5 Air· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75%
123Nemotron 3 Nano 30B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75%
124DeepSeek V3.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.9%
125qwen3-vl-30bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.74.4%
126Gemini 2.0 Flashthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.2%
127GPT OSS 20bhigh· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.2%
128ERNIE 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74%
129Mercury 2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74%
130DeepSeek R1 Zero· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73.3%
131O1 Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73.3%
132LongCat Flash Chat· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73.2%
133Qwen3 VL 32B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.73.1%
134Claude Haiku 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73%
135Qwen3 Next 80B A3B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim72.9%
136GPT OSS 20b· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.5%
137GPT-5 nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.2%
138Ministral 14Blatest· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.2%
139Mistral Smalllatest· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.2%
140Magistral Medium· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim70.8%
141qwen3-vl-30b· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.70.4%
142GPT 4o· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim70.1%
143MiniMax M1 80K· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.70%
144Qwen: Qwen3 VL 8B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.69.9%
145Llama 4 Maverick· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.8%
146GPT-4.5 Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.5%
147MiniMax M1 40K· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.69.2%
148Phi 4 Reasoning Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim68.9%
149Qwen3 VL 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.68.9%
150Magistral Small 25092506· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim68.2%

Showing 101–150 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.