math benchmark
OlympiadBench
A challenging benchmark for promoting AGI with Olympiad-level bilingual multimodal scientific problems. Comprises 8,476 math and physics problems from international and Chinese Olympiads and the Chinese college entrance exam, featuring expert-level annotations for step-by-step reasoning. Includes both text-only and multimodal problems in English and Chinese.
Updated Aug 7, 2026
Published models1
Registry coverage1
MetricScore
EvidenceB
OlympiadBench leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 20.4% | 100.0% | 1 | C |
OlympiadBench highlights
The top published results on this benchmark's own scale.
What is OlympiadBench?
Definition and scoring fields from the benchmark registry.
A challenging benchmark for promoting AGI with Olympiad-level bilingual multimodal scientific problems. Comprises 8,476 math and physics problems from international and Chinese Olympiads and the Chinese college entrance exam, featuring expert-level annotations for step-by-step reasoning. Includes both text-only and multimodal problems in English and Chinese.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- OlympiadBench
- Modality
- multimodal
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- olympiadbench|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about OlympiadBench.
Which model scores highest on OlympiadBench?
QvQ-72B-Preview is currently ranked first with 20.4%.
What does OlympiadBench measure?
A challenging benchmark for promoting AGI with Olympiad-level bilingual multimodal scientific problems. Comprises 8,476 math and physics problems from international and Chinese Olympiads and the Chinese college entrance exam, featuring expert-level annotations for step-by-step reasoning. Includes both text-only and multimodal problems in English and Chinese.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.