Knowledge · Benchmark profile

MMLU-ProX

A multilingual extension of professional-level academic evaluation across many languages.

Reference2025Active12 models

Data verified 21 Jul 2026 · Methodology 1.6.0

Benchmark score on MMLU-ProX

Leaderboard · 12 models

Qwen3.7-Max leads at 87%, followed by Claude Opus 4.5 (85.7%) and Qwen3.7-Plus (85.4%).

Display onlyKnowledge2025

#1Closed

Qwen3.7-Max

Alibaba Cloud

qwen3.7-max

87%

Overall #37 · Context 1M tokens

#2Closed

Claude Opus 4.5

Anthropic

claude-opus-4-5

85.7%

Overall #35 · Context 1M tokens

#3Closed

Qwen3.7-Plus

Alibaba Cloud

qwen3.7-plus

85.4%

Overall #20 · Context 1M tokens

Visual analysis

See the field.

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

ViewModelsColours

Alibaba CloudAnthropicZ.AINVIDIAMoonshot AI

MMLU-ProX leaderboard

12 ranked models · higher is better

View accessible chart data

Accessible data for MMLU-ProX leaderboard
Model	Rank	Provider	Score
Qwen3.7-Max	#1	Alibaba Cloud	87%
Claude Opus 4.5	#2	Anthropic	85.7%
Qwen3.7-Plus	#3	Alibaba Cloud	85.4%
Qwen3.5 397B	#4	Alibaba Cloud	84.7%
Qwen3.6 Plus	#5	Alibaba Cloud	84.7%
GLM-5	#6	Z.AI	83.1%
Nemotron 3 Ultra	#7	NVIDIA	83%
Kimi K2.5	#8	Moonshot AI	82.3%
Qwen3.5 27B	#9	Alibaba Cloud	82.2%
Qwen3.5-122B-A10B	#10	Alibaba Cloud	82.2%
Qwen3.5-35B-A3B	#11	Alibaba Cloud	81%
Qwen3 235B 2507	#12	Alibaba Cloud	79.4%

Benchmark score table (12 models)

One best score per model · higher is better

MMLU-ProX leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	Reference only	87%
#2	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	Reference only	85.7%
#3	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	Reference only	85.4%
#4	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	Estimated reference	84.7%
#5	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	Reference only	84.7%
#6	GLM-5 glm-5	Z.AI	open	Reference only	83.1%
#7	Nemotron 3 Ultra nemotron-3-ultra-550b	NVIDIA	open	Estimated reference	83%
#8	Kimi K2.5 kimi-k2.5	Moonshot AI	—	Reference only	82.3%
#9	Qwen3.5 27B qwen3.5-27b	Alibaba Cloud	open	Reference only	82.2%
#10	Qwen3.5-122B-A10B qwen3-5-122b-a10b	Alibaba Cloud	open	Reference only	82.2%
#11	Qwen3.5-35B-A3B qwen3-5-35b-a3b	Alibaba Cloud	open	Reference only	81%
#12	Qwen3 235B 2507 qwen3-235b-2507	Alibaba Cloud	open	Estimated reference	79.4%

The top of this snapshot is led by Qwen3.7-Max at 87%; third place is 1.6 points behind. The top-10 spread is 4.8 points.

Compare top models

Qwen3.7-Max vs Claude Opus 4.5 Claude Opus 4.5 vs Qwen3.7-Plus Qwen3.7-Plus vs Qwen3.5 397B Qwen3.5 397B vs Qwen3.6 Plus

About MMLU-ProX

Definition and scoring

Organisation: MMLU-ProX authors
Category: Knowledge
Version: 2025
Direction: higher is better
Ranking use: Reference
Contamination risk: Unknown

A multilingual extension of professional-level academic evaluation across many languages. 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 MMLU-ProX measure?

A multilingual extension of professional-level academic evaluation across many languages.

Which model scores highest on MMLU-ProX?

Qwen3.7-Max by Alibaba Cloud currently leads with 87%.

How many models are evaluated on MMLU-ProX?

12 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

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

2025 · 19 results · Reference

Massive Multitask Language Understanding Professional

2024 · 42 results · Reference

Knowledge · Benchmark profile

MMLU-ProX

A multilingual extension of professional-level academic evaluation across many languages.

Reference2025Active12 models

Data verified 21 Jul 2026 · Methodology 1.6.0

Benchmark score on MMLU-ProX

Leaderboard · 12 models

Qwen3.7-Max leads at 87%, followed by Claude Opus 4.5 (85.7%) and Qwen3.7-Plus (85.4%).

Display onlyKnowledge2025

#1Closed

Qwen3.7-Max

Alibaba Cloud

qwen3.7-max

87%

Overall #37 · Context 1M tokens

#2Closed

Claude Opus 4.5

Anthropic

claude-opus-4-5

85.7%

Overall #35 · Context 1M tokens

#3Closed

Qwen3.7-Plus

Alibaba Cloud

qwen3.7-plus

85.4%

Overall #20 · Context 1M tokens

Visual analysis

See the field.

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

ViewModelsColours

Alibaba CloudAnthropicZ.AINVIDIAMoonshot AI

MMLU-ProX leaderboard

12 ranked models · higher is better

View accessible chart data

Accessible data for MMLU-ProX leaderboard
Model	Rank	Provider	Score
Qwen3.7-Max	#1	Alibaba Cloud	87%
Claude Opus 4.5	#2	Anthropic	85.7%
Qwen3.7-Plus	#3	Alibaba Cloud	85.4%
Qwen3.5 397B	#4	Alibaba Cloud	84.7%
Qwen3.6 Plus	#5	Alibaba Cloud	84.7%
GLM-5	#6	Z.AI	83.1%
Nemotron 3 Ultra	#7	NVIDIA	83%
Kimi K2.5	#8	Moonshot AI	82.3%
Qwen3.5 27B	#9	Alibaba Cloud	82.2%
Qwen3.5-122B-A10B	#10	Alibaba Cloud	82.2%
Qwen3.5-35B-A3B	#11	Alibaba Cloud	81%
Qwen3 235B 2507	#12	Alibaba Cloud	79.4%

Benchmark score table (12 models)

One best score per model · higher is better

MMLU-ProX leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	Reference only	87%
#2	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	Reference only	85.7%
#3	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	Reference only	85.4%
#4	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	Estimated reference	84.7%
#5	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	Reference only	84.7%
#6	GLM-5 glm-5	Z.AI	open	Reference only	83.1%
#7	Nemotron 3 Ultra nemotron-3-ultra-550b	NVIDIA	open	Estimated reference	83%
#8	Kimi K2.5 kimi-k2.5	Moonshot AI	—	Reference only	82.3%
#9	Qwen3.5 27B qwen3.5-27b	Alibaba Cloud	open	Reference only	82.2%
#10	Qwen3.5-122B-A10B qwen3-5-122b-a10b	Alibaba Cloud	open	Reference only	82.2%
#11	Qwen3.5-35B-A3B qwen3-5-35b-a3b	Alibaba Cloud	open	Reference only	81%
#12	Qwen3 235B 2507 qwen3-235b-2507	Alibaba Cloud	open	Estimated reference	79.4%

The top of this snapshot is led by Qwen3.7-Max at 87%; third place is 1.6 points behind. The top-10 spread is 4.8 points.

Compare top models

Qwen3.7-Max vs Claude Opus 4.5 Claude Opus 4.5 vs Qwen3.7-Plus Qwen3.7-Plus vs Qwen3.5 397B Qwen3.5 397B vs Qwen3.6 Plus

About MMLU-ProX

Definition and scoring

Organisation: MMLU-ProX authors
Category: Knowledge
Version: 2025
Direction: higher is better
Ranking use: Reference
Contamination risk: Unknown

Open benchmark source ↗

FAQ

Quick answers

What does MMLU-ProX measure?

A multilingual extension of professional-level academic evaluation across many languages.

Which model scores highest on MMLU-ProX?

Qwen3.7-Max by Alibaba Cloud currently leads with 87%.

How many models are evaluated on MMLU-ProX?

12 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

SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines

2025 · 19 results · Reference

Massive Multitask Language Understanding Professional

2024 · 42 results · Reference

Search LUMINA BENCH

Navigation

MMLU-ProX

Leaderboard · 12 models

Qwen3.7-Max

Claude Opus 4.5

Qwen3.7-Plus

See the field.

MMLU-ProX leaderboard

Crop MMLU-ProX leaderboard

Benchmark score table (12 models)

Crop MMLU-ProX benchmark table

Compare top models

Definition and scoring

Quick answers

What does MMLU-ProX measure?

Which model scores highest on MMLU-ProX?

How many models are evaluated on MMLU-ProX?

Does this affect overall Lumina rank?

MMLU-ProX

Leaderboard · 12 models

Qwen3.7-Max

Claude Opus 4.5

Qwen3.7-Plus

See the field.

MMLU-ProX leaderboard

Crop MMLU-ProX leaderboard

Benchmark score table (12 models)

Crop MMLU-ProX benchmark table

Compare top models

Definition and scoring

Quick answers

What does MMLU-ProX measure?

Which model scores highest on MMLU-ProX?

How many models are evaluated on MMLU-ProX?

Does this affect overall Lumina rank?

Leaderboard · 12 models

Qwen3.7-Max

Claude Opus 4.5

Qwen3.7-Plus

See the field.

Benchmark score table (12 models)

Compare top models

Definition and scoring

Quick answers

What does MMLU-ProX measure?

Which model scores highest on MMLU-ProX?

How many models are evaluated on MMLU-ProX?

Does this affect overall Lumina rank?

More Knowledge benchmarks

Leaderboard · 12 models

Qwen3.7-Max

Claude Opus 4.5

Qwen3.7-Plus

See the field.

Benchmark score table (12 models)

Compare top models

Definition and scoring

Quick answers

What does MMLU-ProX measure?

Which model scores highest on MMLU-ProX?

How many models are evaluated on MMLU-ProX?

Does this affect overall Lumina rank?

More Knowledge benchmarks