legal benchmark
MMLU-ProX Leaderboard
Extended version of MMLU-Pro providing additional challenging multiple-choice questions for evaluating language models across diverse academic and professional domains. Built on the foundation of the Massive Multitask Language Understanding benchmark framework.
Updated Aug 17, 2026
Models32
Model coverage32
MetricScore
EvidenceB
MMLU-ProX Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score87.0% | Percentile100.0% | Participants32 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score85.4% | Percentile96.8% | Participants32 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score84.7% | Percentile93.5% | Participants32 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score84.7% | Percentile90.3% | Participants32 | EvidenceC | Evaluated |
| Rank05 | ModelNV | Score83.0% | Percentile87.1% | Participants32 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score82.2% | Percentile83.9% | Participants32 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score82.2% | Percentile80.7% | Participants32 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score81.0% | Percentile77.4% | Participants32 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score81.0% | Percentile74.2% | Participants32 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score80.6% | Percentile71.0% | Participants32 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score79.4% | Percentile67.7% | Participants32 | EvidenceC | Evaluated |
| Rank12 | ModelNV | Score79.4% | Percentile64.5% | Participants32 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score78.7% | Percentile61.3% | Participants32 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score77.8% | Percentile58.1% | Participants32 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score77.2% | Percentile54.8% | Participants32 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score76.7% | Percentile51.6% | Participants32 | EvidenceC | Evaluated |
| Rank17 | ModelAC | Score76.3% | Percentile48.4% | Participants32 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score76.1% | Percentile45.2% | Participants32 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score73.4% | Percentile41.9% | Participants32 | EvidenceC | Evaluated |
| Rank20 | ModelAC | Score71.5% | Percentile38.7% | Participants32 | EvidenceC | Evaluated |
| Rank21 | ModelAC | Score70.9% | Percentile35.5% | Participants32 | EvidenceC | Evaluated |
| Rank22 | ModelAC | Score70.7% | Percentile32.3% | Participants32 | EvidenceC | Evaluated |
| Rank23 | ModelAC | Score65.4% | Percentile29.0% | Participants32 | EvidenceC | Evaluated |
| Rank24 | ModelAC | Score65.0% | Percentile25.8% | Participants32 | EvidenceC | Evaluated |
| Rank25 | ModelNV | Score59.5% | Percentile22.6% | Participants32 | EvidenceC | Evaluated |
| Rank26 | ModelAC | Score59.4% | Percentile19.4% | Participants32 | EvidenceC | Evaluated |
| Rank27 | ModelAC | Score52.3% | Percentile16.1% | Participants32 | EvidenceC | Evaluated |
| Rank28 | ModelAC | Score34.6% | Percentile12.9% | Participants32 | EvidenceC | Evaluated |
| Rank29 | ModelGO | Score19.9% | Percentile9.7% | Participants32 | EvidenceC | Evaluated |
| Rank30 | ModelGO | Score19.9% | Percentile6.5% | Participants32 | EvidenceC | Evaluated |
| Rank31 | ModelGO | Score8.1% | Percentile3.2% | Participants32 | EvidenceC | Evaluated |
| Rank32 | ModelGO | Score8.1% | Percentile0.0% | Participants32 | EvidenceC | Evaluated |
MMLU-ProX Highlights
The leading models and scores on this benchmark.
MMLU-ProX Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MMLU-ProX
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis mmlu-prox AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ACQwen3.7 MaxAlibaba Cloud / Qwen Team
- Score
- 87.0%
- Price
- $2.5 input / $7.5 output per 1M tokens
- Speed
- Up to 5.8 tok/s via Together
Strengths
- Ranks #1 of 32 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMLU-ProX, not total model capability
- 02ACQwen3.7-PlusAlibaba Cloud / Qwen Team
- Score
- 85.4%
- Price
- $0.50 input / $3.0 output per 1M tokens
Strengths
- Ranks #2 of 32 compared models
- 97th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMLU-ProX, not total model capability
- 03ACQwen3.5-397B-A17BAlibaba Cloud / Qwen Team
- Score
- 84.7%
- Price
- $0.60 input / $3.6 output per 1M tokens
Strengths
- Ranks #3 of 32 compared models
- 94th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMLU-ProX, not total model capability
- 04ACQwen3.6 PlusAlibaba Cloud / Qwen Team
- Score
- 84.7%
- Price
- $0.50 input / $3.0 output per 1M tokens
- Speed
- Up to 16 tok/s via Together
Strengths
- Ranks #4 of 32 compared models
- 90th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMLU-ProX, not total model capability
- 05NV
- Score
- 83.0%
- Price
- $0.50 input / $2.5 output per 1M tokens
Strengths
- Ranks #5 of 32 compared models
- 87th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMLU-ProX, not total model capability
Selection summary
Best AI Models for MMLU-ProX
Qwen3.7 Max currently leads MMLU-ProX with 87.0%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is MMLU-ProX?
What MMLU-ProX measures and how its scores work.
Extended version of MMLU-Pro providing additional challenging multiple-choice questions for evaluating language models across diverse academic and professional domains. Built on the foundation of the Massive Multitask Language Understanding benchmark framework.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MMLU-ProX
- Modality
- text
- Primary category
- legal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- mmlu-prox|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MMLU-ProX.
Which model scores highest on MMLU-ProX?
Qwen3.7 Max is currently ranked first with 87.0%.
What does MMLU-ProX measure?
Extended version of MMLU-Pro providing additional challenging multiple-choice questions for evaluating language models across diverse academic and professional domains. Built on the foundation of the Massive Multitask Language Understanding benchmark framework.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
32 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.