legal benchmark
AGIEval Leaderboard
A human-centric benchmark for evaluating foundation models on standardized exams including college entrance exams (Gaokao, SAT), law school admission tests (LSAT), math competitions, lawyer qualification tests, and civil service exams. Contains 20 tasks (18 multiple-choice, 2 cloze) designed to assess understanding, knowledge, reasoning, and calculation abilities in real-world academic and professional contexts.
Updated Aug 17, 2026
Models10
Model coverage10
MetricScore
EvidenceB
AGIEval Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMA | Score65.8% | Percentile100.0% | Participants10 | EvidenceC | Evaluated |
| Rank02 | ModelMA | Score64.8% | Percentile88.9% | Participants10 | EvidenceC | Evaluated |
| Rank03 | ModelMA | Score59.1% | Percentile77.8% | Participants10 | EvidenceC | Evaluated |
| Rank04 | ModelNR | Score56.2% | Percentile66.7% | Participants10 | EvidenceC | Evaluated |
| Rank05 | ModelGO | Score55.1% | Percentile55.6% | Participants10 | EvidenceC | Evaluated |
| Rank06 | ModelGO | Score52.8% | Percentile44.4% | Participants10 | EvidenceC | Evaluated |
| Rank07 | ModelMA | Score51.1% | Percentile33.3% | Participants10 | EvidenceC | Evaluated |
| Rank08 | ModelIB | Score49.3% | Percentile22.2% | Participants10 | EvidenceC | Evaluated |
| Rank09 | ModelMA | Score48.3% | Percentile11.1% | Participants10 | EvidenceC | Evaluated |
| Rank10 | ModelBA | Score28.5% | Percentile0.0% | Participants10 | EvidenceC | Evaluated |
AGIEval Highlights
The leading models and scores on this benchmark.
AGIEval Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for AGIEval
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis agieval 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.
- 01MAMistral Small 3 24B BaseMistral AI
- Score
- 65.8%
- Speed
- Up to 134 tok/s via Mistral AI
Strengths
- Ranks #1 of 10 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures AGIEval, not total model capability
- 02MAMinistral 3 (14B Base 2512)Mistral AI
- Score
- 64.8%
- Price
- $0.10 input / $0.10 output per 1M tokens
- Speed
- Up to 238 tok/s via Mistral AI
Strengths
- Ranks #2 of 10 compared models
- 89th percentile on this benchmark
- C evidence result
Considerations
- This result measures AGIEval, not total model capability
- 03MAMinistral 3 (8B Base 2512)Mistral AI
- Score
- 59.1%
- Price
- $0.10 input / $0.10 output per 1M tokens
- Speed
- Up to 238 tok/s via Mistral AI
Strengths
- Ranks #3 of 10 compared models
- 78th percentile on this benchmark
- C evidence result
Considerations
- This result measures AGIEval, not total model capability
- 04NRHermes 3 70BNous Research
- Score
- 56.2%
Strengths
- Ranks #4 of 10 compared models
- 67th percentile on this benchmark
- C evidence result
Considerations
- This result measures AGIEval, not total model capability
- 05GOGemma 2 27BGoogle
- Score
- 55.1%
Strengths
- Ranks #5 of 10 compared models
- 56th percentile on this benchmark
- C evidence result
Considerations
- This result measures AGIEval, not total model capability
Selection summary
Best AI Models for AGIEval
Mistral Small 3 24B Base currently leads AGIEval with 65.8%. 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 AGIEval?
What AGIEval measures and how its scores work.
A human-centric benchmark for evaluating foundation models on standardized exams including college entrance exams (Gaokao, SAT), law school admission tests (LSAT), math competitions, lawyer qualification tests, and civil service exams. Contains 20 tasks (18 multiple-choice, 2 cloze) designed to assess understanding, knowledge, reasoning, and calculation abilities in real-world academic and professional contexts.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- AGIEval
- Modality
- text
- Primary category
- legal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- agieval|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about AGIEval.
Which model scores highest on AGIEval?
Mistral Small 3 24B Base is currently ranked first with 65.8%.
What does AGIEval measure?
A human-centric benchmark for evaluating foundation models on standardized exams including college entrance exams (Gaokao, SAT), law school admission tests (LSAT), math competitions, lawyer qualification tests, and civil service exams. Contains 20 tasks (18 multiple-choice, 2 cloze) designed to assess understanding, knowledge, reasoning, and calculation abilities in real-world academic and professional contexts.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
10 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.