SuperGPQA

Multilingual: no

SuperGPQA is a comprehensive benchmark that evaluates large language models across 285 graduate-level academic disciplines. The benchmark contains 25,957 questions covering 13 broad disciplinary areas including Engineering, Medicine, Science, and Law, with specialized fields in light industry, agriculture, and service-oriented domains. It employs a Human-LLM collaborative filtering mechanism with over 80 expert annotators to create challenging questions that assess graduate-level knowledge and reasoning capabilities.

Methodology

Imported from llm-stats public benchmark metadata. Modality: text. Max score: 1. Categories: chemistry, economics, finance, general, healthcare, legal, math, physics, reasoning. Language: en. Verified by llm-stats: no.

Leaderboard

Showing 20 of 25 results

Qwen3.7 Max self-reported llm-stats

73.6%
Qwen3.6 Plus self-reported llm-stats

71.6%
Qwen3.5-397B-A17B self-reported llm-stats

70.4%
Qwen3.5-122B-A10B self-reported llm-stats

67.1%
Qwen3.6-27B self-reported llm-stats

66.0%
Qwen3.5-27B self-reported llm-stats

65.6%
Qwen3-235B-A22B-Thinking-2507 self-reported llm-stats

64.9%
Qwen3 VL 235B A22B Thinking self-reported llm-stats

64.3%
Qwen3.5-35B-A3B self-reported llm-stats

63.4%
Qwen3-235B-A22B-Instruct-2507 self-reported llm-stats

62.6%
Qwen3-Next-80B-A3B-Thinking self-reported llm-stats

60.8%
Qwen3 VL 235B A22B Instruct self-reported llm-stats

60.4%
Qwen3 VL 32B Thinking self-reported llm-stats

59.0%
Qwen3-Next-80B-A3B-Instruct self-reported llm-stats

58.8%
Kimi K2 Instruct self-reported llm-stats

57.2%
Kimi K2-Instruct-0905 self-reported llm-stats

57.2%
Qwen3 VL 30B A3B Thinking self-reported llm-stats

56.4%
Qwen3 VL 32B Instruct self-reported llm-stats

54.6%
Qwen3 VL 30B A3B Instruct self-reported llm-stats

53.1%
Qwen3 VL 8B Thinking self-reported llm-stats

51.2%