math benchmark
PhysicsFinals
PHYSICS is a comprehensive benchmark for university-level physics problem solving, containing 1,297 expert-annotated problems covering six core areas: classical mechanics, quantum mechanics, thermodynamics and statistical mechanics, electromagnetism, atomic physics, and optics. Each problem requires advanced physics knowledge and mathematical reasoning. Even advanced models like o3-mini achieve only 59.9% accuracy.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
PhysicsFinals leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | GO | 63.9% | 100.0% | 2 | C | |
| 2 | GO | 57.4% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
PhysicsFinals highlights
The top published results on this benchmark's own scale.
What is PhysicsFinals?
Definition and scoring fields from the benchmark registry.
PHYSICS is a comprehensive benchmark for university-level physics problem solving, containing 1,297 expert-annotated problems covering six core areas: classical mechanics, quantum mechanics, thermodynamics and statistical mechanics, electromagnetism, atomic physics, and optics. Each problem requires advanced physics knowledge and mathematical reasoning. Even advanced models like o3-mini achieve only 59.9% accuracy.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- PhysicsFinals
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- physicsfinals|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about PhysicsFinals.
Which model scores highest on PhysicsFinals?
Gemini 1.5 Pro is currently ranked first with 63.9%.
What does PhysicsFinals measure?
PHYSICS is a comprehensive benchmark for university-level physics problem solving, containing 1,297 expert-annotated problems covering six core areas: classical mechanics, quantum mechanics, thermodynamics and statistical mechanics, electromagnetism, atomic physics, and optics. Each problem requires advanced physics knowledge and mathematical reasoning. Even advanced models like o3-mini achieve only 59.9% accuracy.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 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.