MATH Level 5
Competition mathematics problems at the dataset's hardest difficulty band, scored on the final answer.
contamination: mediummetric: accuracy
Comparable set
math-level-5/v1/level-5/0shot/exact-match/inspect
| Model | Score | Uncertainty | Run by | Retrieved |
|---|---|---|---|---|
| DeepSeek-R1-0528 | 96.6% | unknown | third-party | 2026-08-10 |
| gemma-3-27b-it | 74.0% | ±0.0112 | third-party | 2026-08-10 |
| Llama-4-Maverick-17B-128E-Instruct | 73.0% | unknown | third-party | 2026-08-10 |
| Qwen3-235B-A22B-Instruct-2507 | 68.9% | ±0.0095 | third-party | 2026-08-10 |
| phi-4 | 64.9% | unknown | third-party | 2026-08-10 |
| Qwen2.5-72B-Instruct | 63.2% | unknown | third-party | 2026-08-10 |
| Llama-4-Scout-17B-16E-Instruct | 62.3% | unknown | third-party | 2026-08-10 |
| Llama-3.3-70B-Instruct | 41.6% | unknown | third-party | 2026-08-10 |
| Llama-3.1-70B-Instruct | 36.7% | unknown | third-party | 2026-08-10 |
| Llama-3.1-8B-Instruct | 22.9% | unknown | third-party | 2026-08-10 |
| gemma-2-9b-it | 21.0% | unknown | third-party | 2026-08-10 |
| Mistral-Nemo-Instruct-2407 | 10.8% | ±0.0061 | third-party | 2026-08-10 |
Epoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. · license CC-BY-4.0 · epoch-ai · https://epoch.ai/benchmarks