Knowledge · Benchmark profile

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

An expanded version of GPQA that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines, providing comprehensive coverage of academic domains.

Reference2025Active19 models

Data verified 21 Jul 2026 · Methodology 1.6.0

Benchmark score on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

Leaderboard · 19 models

Claude Opus 4.6 leads at 95%, followed by Claude Sonnet 4.6 (95%) and Qwen 3.6 Max (preview) (73.9%).

Display onlyKnowledge2025

#1Closed

Claude Opus 4.6

Anthropic

claude-opus-4-6

95%

Overall #17 · Context 1M tokens

#2Closed

Claude Sonnet 4.6

Anthropic

claude-sonnet-4-6

95%

Overall #27 · Context 1M tokens

#3Closed

Qwen 3.6 Max (preview)

Alibaba Cloud

qwen3-6-max-preview

73.9%

Overall — · Context 256K tokens

Visual analysis

See the field.

Switch between model placement, score distribution and descriptive provider averages. Every view uses the same sourced leaderboard.

ViewModelsColours

AnthropicAlibaba CloudMoonshot AIZ.AIDeepSeekOpenBMB

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard

19 ranked models · higher is better

View accessible chart data

Accessible data for SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard
Model	Rank	Provider	Score
Claude Opus 4.6	#1	Anthropic	95%
Claude Sonnet 4.6	#2	Anthropic	95%
Qwen 3.6 Max (preview)	#3	Alibaba Cloud	73.9%
Qwen3.7-Max	#4	Alibaba Cloud	73.6%
Qwen3.6 Plus	#5	Alibaba Cloud	71.6%
Qwen3.7-Plus	#6	Alibaba Cloud	71.4%
Claude Opus 4.5	#7	Anthropic	70.6%
Qwen3.5 397B	#8	Alibaba Cloud	70.4%
Kimi K2.5	#9	Moonshot AI	69.2%
Qwen3.5-122B-A10B	#10	Alibaba Cloud	67.1%
GLM-5	#11	Z.AI	66.8%
Qwen3.6 27B	#12	Alibaba Cloud	66%
Qwen3.5 27B	#13	Alibaba Cloud	65.6%
Qwen3.6-35B-A3B	#14	Alibaba Cloud	64.7%
Qwen3.5-35B-A3B	#15	Alibaba Cloud	63.4%
Qwen3 235B 2507	#16	Alibaba Cloud	62.6%
DeepSeek V4 Pro Base	#17	DeepSeek	53.9%
DeepSeek V4 Flash Base	#18	DeepSeek	46.5%
MiniCPM5-1B	#19	OpenBMB	23.1%

Benchmark score table (19 models)

One best score per model · higher is better

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	Reference only	95%
#2	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	Reference only	95%
#3	Qwen 3.6 Max (preview) qwen3-6-max-preview	Alibaba Cloud	closed	Reference only	73.9%
#4	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	Reference only	73.6%
#5	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	Reference only	71.6%
#6	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	Reference only	71.4%
#7	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	Reference only	70.6%
#8	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	Estimated reference	70.4%
#9	Kimi K2.5 kimi-k2.5	Moonshot AI	—	Reference only	69.2%
#10	Qwen3.5-122B-A10B qwen3-5-122b-a10b	Alibaba Cloud	open	Reference only	67.1%
#11	GLM-5 glm-5	Z.AI	open	Reference only	66.8%
#12	Qwen3.6 27B qwen3.6-27b	Alibaba Cloud	open	Estimated reference	66%
#13	Qwen3.5 27B qwen3.5-27b	Alibaba Cloud	open	Reference only	65.6%
#14	Qwen3.6-35B-A3B qwen3-6-35b-a3b	Alibaba Cloud	open	Estimated reference	64.7%
#15	Qwen3.5-35B-A3B qwen3-5-35b-a3b	Alibaba Cloud	open	Reference only	63.4%
#16	Qwen3 235B 2507 qwen3-235b-2507	Alibaba Cloud	open	Estimated reference	62.6%
#17	DeepSeek V4 Pro Base deepseek-v4-pro-base	DeepSeek	open	Estimated reference	53.9%
#18	DeepSeek V4 Flash Base deepseek-v4-flash-base	DeepSeek	open	Estimated reference	46.5%
#19	MiniCPM5-1B minicpm5-1b	OpenBMB	open	Estimated reference	23.1%

The top of this snapshot is led by Claude Opus 4.6 at 95%; third place is 21.1 points behind. The top-10 spread is 27.9 points.

Compare top models

Claude Opus 4.6 vs Claude Sonnet 4.6 Claude Sonnet 4.6 vs Qwen 3.6 Max (preview)Qwen 3.6 Max (preview) vs Qwen3.7-Max Qwen3.7-Max vs Qwen3.6 Plus

About SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

Definition and scoring

Organisation: Xiaoxuan Du, Yao Yao, Kexin Ma, Bowen Wang, Tianyu Zheng, Kaiyan Zhu, Yiming Zhang, Yutao Zhu, Jiawei Zhou, Jingren Zhou
Category: Knowledge
Version: 2025
Direction: higher is better
Ranking use: Reference
Contamination risk: Unknown

An expanded version of GPQA that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines, providing comprehensive coverage of academic domains. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.

Open benchmark source ↗

FAQ

Quick answers

What does SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines measure?

An expanded version of GPQA that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines, providing comprehensive coverage of academic domains.

Which model scores highest on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

Claude Opus 4.6 by Anthropic currently leads with 95%.

How many models are evaluated on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

19 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

No. This benchmark is display-only and does not enter the overall Lumina composite.

Artificial Analysis Openness Index

2026 · 10 results · Reference

Artificial Analysis MMLU-Pro

2026 · 4 results · Reference

Massive Multitask Language Understanding

2020 · 8 results · Reference

Graduate-Level Google-Proof Q&A

2023 · 70 results · Reference

Massive Multitask Language Understanding Professional

2024 · 42 results · Reference

AGIEval

2026 · 3 results · Reference

Knowledge · Benchmark profile

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

An expanded version of GPQA that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines, providing comprehensive coverage of academic domains.

Reference2025Active19 models

Data verified 21 Jul 2026 · Methodology 1.6.0

Benchmark score on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

Leaderboard · 19 models

Claude Opus 4.6 leads at 95%, followed by Claude Sonnet 4.6 (95%) and Qwen 3.6 Max (preview) (73.9%).

Display onlyKnowledge2025

#1Closed

Claude Opus 4.6

Anthropic

claude-opus-4-6

95%

Overall #17 · Context 1M tokens

#2Closed

Claude Sonnet 4.6

Anthropic

claude-sonnet-4-6

95%

Overall #27 · Context 1M tokens

#3Closed

Qwen 3.6 Max (preview)

Alibaba Cloud

qwen3-6-max-preview

73.9%

Overall — · Context 256K tokens

Visual analysis

See the field.

Switch between model placement, score distribution and descriptive provider averages. Every view uses the same sourced leaderboard.

ViewModelsColours

AnthropicAlibaba CloudMoonshot AIZ.AIDeepSeekOpenBMB

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard

19 ranked models · higher is better

View accessible chart data

Accessible data for SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard
Model	Rank	Provider	Score
Claude Opus 4.6	#1	Anthropic	95%
Claude Sonnet 4.6	#2	Anthropic	95%
Qwen 3.6 Max (preview)	#3	Alibaba Cloud	73.9%
Qwen3.7-Max	#4	Alibaba Cloud	73.6%
Qwen3.6 Plus	#5	Alibaba Cloud	71.6%
Qwen3.7-Plus	#6	Alibaba Cloud	71.4%
Claude Opus 4.5	#7	Anthropic	70.6%
Qwen3.5 397B	#8	Alibaba Cloud	70.4%
Kimi K2.5	#9	Moonshot AI	69.2%
Qwen3.5-122B-A10B	#10	Alibaba Cloud	67.1%
GLM-5	#11	Z.AI	66.8%
Qwen3.6 27B	#12	Alibaba Cloud	66%
Qwen3.5 27B	#13	Alibaba Cloud	65.6%
Qwen3.6-35B-A3B	#14	Alibaba Cloud	64.7%
Qwen3.5-35B-A3B	#15	Alibaba Cloud	63.4%
Qwen3 235B 2507	#16	Alibaba Cloud	62.6%
DeepSeek V4 Pro Base	#17	DeepSeek	53.9%
DeepSeek V4 Flash Base	#18	DeepSeek	46.5%
MiniCPM5-1B	#19	OpenBMB	23.1%

Benchmark score table (19 models)

One best score per model · higher is better

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	Reference only	95%
#2	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	Reference only	95%
#3	Qwen 3.6 Max (preview) qwen3-6-max-preview	Alibaba Cloud	closed	Reference only	73.9%
#4	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	Reference only	73.6%
#5	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	Reference only	71.6%
#6	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	Reference only	71.4%
#7	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	Reference only	70.6%
#8	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	Estimated reference	70.4%
#9	Kimi K2.5 kimi-k2.5	Moonshot AI	—	Reference only	69.2%
#10	Qwen3.5-122B-A10B qwen3-5-122b-a10b	Alibaba Cloud	open	Reference only	67.1%
#11	GLM-5 glm-5	Z.AI	open	Reference only	66.8%
#12	Qwen3.6 27B qwen3.6-27b	Alibaba Cloud	open	Estimated reference	66%
#13	Qwen3.5 27B qwen3.5-27b	Alibaba Cloud	open	Reference only	65.6%
#14	Qwen3.6-35B-A3B qwen3-6-35b-a3b	Alibaba Cloud	open	Estimated reference	64.7%
#15	Qwen3.5-35B-A3B qwen3-5-35b-a3b	Alibaba Cloud	open	Reference only	63.4%
#16	Qwen3 235B 2507 qwen3-235b-2507	Alibaba Cloud	open	Estimated reference	62.6%
#17	DeepSeek V4 Pro Base deepseek-v4-pro-base	DeepSeek	open	Estimated reference	53.9%
#18	DeepSeek V4 Flash Base deepseek-v4-flash-base	DeepSeek	open	Estimated reference	46.5%
#19	MiniCPM5-1B minicpm5-1b	OpenBMB	open	Estimated reference	23.1%

The top of this snapshot is led by Claude Opus 4.6 at 95%; third place is 21.1 points behind. The top-10 spread is 27.9 points.

Compare top models

Claude Opus 4.6 vs Claude Sonnet 4.6 Claude Sonnet 4.6 vs Qwen 3.6 Max (preview)Qwen 3.6 Max (preview) vs Qwen3.7-Max Qwen3.7-Max vs Qwen3.6 Plus

About SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

Definition and scoring

Organisation: Xiaoxuan Du, Yao Yao, Kexin Ma, Bowen Wang, Tianyu Zheng, Kaiyan Zhu, Yiming Zhang, Yutao Zhu, Jiawei Zhou, Jingren Zhou
Category: Knowledge
Version: 2025
Direction: higher is better
Ranking use: Reference
Contamination risk: Unknown

Open benchmark source ↗

FAQ

Quick answers

What does SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines measure?

An expanded version of GPQA that evaluates graduate-level knowledge and reasoning capabilities across 285 disciplines, providing comprehensive coverage of academic domains.

Which model scores highest on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

Claude Opus 4.6 by Anthropic currently leads with 95%.

How many models are evaluated on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

19 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

No. This benchmark is display-only and does not enter the overall Lumina composite.

Artificial Analysis Openness Index

2026 · 10 results · Reference

Artificial Analysis MMLU-Pro

2026 · 4 results · Reference

Massive Multitask Language Understanding

2020 · 8 results · Reference

Graduate-Level Google-Proof Q&A

2023 · 70 results · Reference

Massive Multitask Language Understanding Professional

2024 · 42 results · Reference

AGIEval

2026 · 3 results · Reference

Search LUMINA BENCH

Navigation

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

Leaderboard · 19 models

Claude Opus 4.6

Claude Sonnet 4.6

Qwen 3.6 Max (preview)

See the field.

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard

Crop SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard

Benchmark score table (19 models)

Crop SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines benchmark table

Compare top models

Definition and scoring

Quick answers

What does SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines measure?

Which model scores highest on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

How many models are evaluated on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

Does this affect overall Lumina rank?

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

Leaderboard · 19 models

Claude Opus 4.6

Claude Sonnet 4.6

Qwen 3.6 Max (preview)

See the field.

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard

Crop SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines leaderboard

Benchmark score table (19 models)

Crop SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines benchmark table

Compare top models

Definition and scoring

Quick answers

What does SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines measure?

Which model scores highest on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

How many models are evaluated on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

Does this affect overall Lumina rank?

Leaderboard · 19 models

Claude Opus 4.6

Claude Sonnet 4.6

Qwen 3.6 Max (preview)

See the field.

Benchmark score table (19 models)

Compare top models

Definition and scoring

Quick answers

What does SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines measure?

Which model scores highest on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

How many models are evaluated on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

Does this affect overall Lumina rank?

More Knowledge benchmarks

Leaderboard · 19 models

Claude Opus 4.6

Claude Sonnet 4.6

Qwen 3.6 Max (preview)

See the field.

Benchmark score table (19 models)

Compare top models

Definition and scoring

Quick answers

What does SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines measure?

Which model scores highest on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

How many models are evaluated on SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines?

Does this affect overall Lumina rank?

More Knowledge benchmarks