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
351Qwen2.5 Math 7B Instruct· Hugging Face Open LLM Leaderboard data1.6%
352NexusRaven-V2-13B· Hugging Face Open LLM Leaderboard data1.3%
353Llama 3 8B Instruct· Hugging Face Open LLM Leaderboard data1.2%
354Qwen2.5-Coder-7B· Hugging Face Open LLM Leaderboard data1.2%
355Llama 3.1 Nemotron 70B Instruct HF· Hugging Face Open LLM Leaderboard data1.1%
356vicuna-7b-v1.5· Hugging Face Open LLM Leaderboard data1.1%
357Qwen2.5 0.5b Instruct· Hugging Face Open LLM Leaderboard data1%
358Qwen2.5 1.5B Instruct· Hugging Face Open LLM Leaderboard data0.783%
359DeepSeek R1 Distill Llama 8B· Hugging Face Open LLM Leaderboard data0.671%
360gemma-2b· Hugging Face Open LLM Leaderboard data0.671%
361RedPajama-INCITE-7B-Base· Hugging Face Open LLM Leaderboard data0.671%
362Qwen1.5-0.5B· Hugging Face Open LLM Leaderboard data0.559%
363Llama-2-7b-chat-hf· Hugging Face Open LLM Leaderboard data0.447%
364RedPajama-INCITE-7B-Chat· Hugging Face Open LLM Leaderboard data0.336%
365Llama-2-7B-32K-Instruct· Hugging Face Open LLM Leaderboard data0.224%
366RedPajama-INCITE-7B-Instruct· Hugging Face Open LLM Leaderboard data0.112%
367Llama 3.2 1B· Hugging Face Open LLM Leaderboard data0%
368Llama-2-13b-chat-hf· Hugging Face Open LLM Leaderboard data0%
369Mistral (7B) Instruct v0.1· Hugging Face Open LLM Leaderboard data0%
370qwen2-0.5b-instruct· Hugging Face Open LLM Leaderboard data0%
371recurrentgemma-2b· Hugging Face Open LLM Leaderboard data0%
372RedPajama-INCITE-Base-3B-v1· Hugging Face Open LLM Leaderboard data0%
373RedPajama-INCITE-Chat-3B-v1· Hugging Face Open LLM Leaderboard data0%
374RedPajama-INCITE-Instruct-3B-v1· Hugging Face Open LLM Leaderboard data0%

Showing 351–374 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.