Benchmark
MathArena Project Euler
Math benchmarks measure how well models solve maths problems. Scores are percentages.
All math benchmarksScores from MathArena data.
- Models measured
- 17
- Scores recorded
- 17
- Sources
- 1
- IQ weight
- 1.38
Score distribution
- 62.6%
- Mean
- 61.5%
- Median
- 17.6%
- Std dev
- 46.1%
- 10th percentile
- 87.7%
- 90th percentile
- 26.4%
- Min
This benchmark has a weight of 1.38 for chat models, measured over 17 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | |
|---|---|---|
| 1 | Gemini 3.1 Pro Previewhigh· MathArena data | 89% |
| 2 | GPT 5.4xhigh· MathArena data | 89% |
| 3 | Claude Opus 4.6high· MathArena data | 87.7% |
| 4 | Gemini 3.5 Flashhigh· MathArena data | 82% |
| 5 | GPT 5.2high· MathArena data | 81.6% |
| 6 | GLM 5.1· MathArena data | 67.2% |
| 7 | Gemini 3 Pro Previewpreview· MathArena data | 62.4% |
| 8 | Gemini 3 Flashhigh· MathArena data | 61.8% |
| 9 | GPT 5.1high· MathArena data | 61.5% |
| 10 | Kimi K2.5high· MathArena data | 59.9% |
| 11 | GPT 5high· MathArena data | 56.9% |
| 12 | DeepSeek V3.2think· MathArena data | 50% |
| 13 | Kimi K2 Thinking· MathArena data | 49.8% |
| 14 | o4 Minihigh· MathArena data | 48% |
| 15 | Grok 4· MathArena data | 46.1% |
| 16 | Grok 4.1 Fastreasoning· MathArena data | 44.9% |
| 17 | Gemini 2.5 Pro· MathArena data | 26.4% |
Tokens the model generated per task, as measured by the source. Comparable within this benchmark only.
Data credits: MathArena (ETH SRI / INSAIT), MIT — cite dekoninck2026matharena (licence)