# AIME 2024 → MATH-500

Calibration · https://benchgap.net/calibration/aime2024/current/to/math-500/current · data retrieved 2026-11-18

MATH-500 is estimated from AIME 2024 with a Michaelis–Menten curve fitted on 13 models measured on both: y = 1.0449·x / (0.06006 + x), R² = 0.96, cross-validated error 1.2 pp. It is used for 12 estimates.

| Estimated model | AIME 2024 | MATH-500 | Source |
|---|---|---|---|
| [Ripple-65B Preview](https://benchgap.net/model/ripple-65b-preview.md) | 97.1% | 98.4% | estimated ± 1.2 pp, medium confidence |
| [GPT-OSS 120B](https://benchgap.net/model/gpt-oss-120b.md) | 95.8% | 98.3% | estimated ± 1.2 pp, medium confidence |
| [GPT-OSS 20B](https://benchgap.net/model/gpt-oss-20b.md) | 92.1% | 98.1% | estimated ± 1.2 pp, medium confidence |
| [Gemini 2.5 Pro](https://benchgap.net/model/gemini-2-5-pro.md) | 92.0% | 98.1% | estimated ± 1.2 pp, medium confidence |
| [DeepSeek-R1-0528](https://benchgap.net/model/deepseek-r1-0528.md) | 91.4% | 98.0% | estimated ± 1.2 pp, medium confidence |
| [Grok 3 Mini](https://benchgap.net/model/grok-3-mini.md) | 89.5% | 97.9% | estimated ± 1.2 pp, medium confidence |
| [o3-mini](https://benchgap.net/model/o3-mini.md) | 87.3% | 97.8% | estimated ± 1.2 pp, medium confidence |
| [Phi-4-reasoning-plus](https://benchgap.net/model/phi-4-reasoning-plus.md) | 81.3% | 97.3% | estimated ± 1.2 pp, high confidence |
| [Phi-4-reasoning](https://benchgap.net/model/phi-4-reasoning.md) | 75.3% | 96.8% | estimated ± 1.2 pp, high confidence |
| [DeepSeek V3.1](https://benchgap.net/model/deepseek-v3-1.md) | 66.3% | 95.8% | estimated ± 1.2 pp, high confidence |
| [Grok 3 [Beta]](https://benchgap.net/model/grok-3-beta.md) | 52.2% | 93.7% | estimated ± 1.2 pp, high confidence |
| [Claude 3.5 Haiku](https://benchgap.net/model/claude-3-5-haiku.md) | 5.3% | 49.0% | estimated ± 1.2 pp, medium confidence |

## Fitted on

| Model | AIME 2024 | MATH-500 |
|---|---|---|
| [Claude 3.7 Sonnet](https://benchgap.net/model/claude-3-7-sonnet.md) | 23.3% | 82.2% |
| [Claude 3.7 Sonnet Thinking](https://benchgap.net/model/claude-3-7-sonnet-thinking.md) | 61.3% | 96.2% |
| [DeepSeek-R1](https://benchgap.net/model/deepseek-r1.md) | 79.8% | 97.3% |
| [DeepSeek V3](https://benchgap.net/model/deepseek-v3.md) | 39.2% | 90.2% |
| [Kimi K2](https://benchgap.net/model/kimi-k2.md) | 69.6% | 97.4% |
| [Magistral Medium](https://benchgap.net/model/magistral-medium.md) | 73.6% | 94.3% |
| [Magistral Small](https://benchgap.net/model/magistral-small.md) | 70.7% | 95.9% |
| [MiniMax M1 80k](https://benchgap.net/model/minimax-m1-80k.md) | 86.0% | 96.8% |
| [Qwen3 235B-A22B](https://benchgap.net/model/qwen3-235b-a22b.md) | 40.1% | 91.2% |
| [Qwen3 235B-A22B (Reasoning)](https://benchgap.net/model/qwen3-235b-a22b-reasoning.md) | 85.7% | 98.0% |
| [Qwen3 32B](https://benchgap.net/model/qwen3-32b.md) | 31.0% | 88.6% |
| [Qwen3 32B (Reasoning)](https://benchgap.net/model/qwen3-32b-reasoning.md) | 81.4% | 97.2% |
| [QwQ-32B](https://benchgap.net/model/qwq-32b.md) | 79.5% | 98.0% |
