Benchmarks · math
MATH 500 (Vals)
Academic math benchmark on probability, algebra, and trigonometry Scores are the values published by VALS AI. Source benchmark version 1.
7 models · 7 results · updated 2026-01-09 · vals.ai ↗
Leaderboard
gpt-oss-120b high
Heavy line: best score at each price. Thin lines join one model's effort levels. Cost uses each model's median subscription plan, or API price with no plan data. 3 more models under All models in settings.
More from this benchmark
1 published measuresTime per task60 results · lower is better
4 more models under All models in settings.
All results
7 rows| # | Model | Effort | MATH 500 (Vals) | Subscription $ / task | API $ / task | Time |
|---|---|---|---|---|---|---|
| 1 | gpt-oss-120b | high | 94.8% | $0.001 | $0.001 | 16s |
| 2 | gpt-oss-20b | high | 94.2% | $0.001 | $0.001 | 31s |
| 3 | GLM 4.5 | xhigh | 94.0% | $0.013 | $0.013 | 85s |
| 4 | DeepSeek R1 | xhigh | 92.2% | $0.011 | $0.011 | 2.6 min |
| 5 | Grok 3 | none | 89.8% | $0.013 | $0.013 | 9.1s |
| 6 | MiniMax M2.1 | 89.0% | $0.005 | $0.005 | 60s | |
| 7 | DeepSeek V3 | none | 80.4% | $0.001 | $0.001 | 7.9s |