LiveBench · Mathematics
The Frontier
Best score over time · one chart, every benchmark
Distribution
Where models cluster
Correlated benchmarks
Pearson r · original research
Benchmarks that track with LiveBench · Mathematics
Pearson correlation across models scored on both benchmarks. Closer to 1 = strongly predictive.
Full rankings
29 models tested · sorted by score
| # | Model | Score |
|---|---|---|
| 1 |  GPT-5.2-Codex | 88.8 |
| 2 |  GLM 5.1 | 84.9 |
| 3 |  Qwen3.6 Plus | 83.7 |
| 4 | GPT-5.1-Codex-Max | 83.7 |
| 5 | GLM 5 | 83.5 |
| 6 | GLM 4.6 | 81.1 |
| 7 |  Kimi K2 Thinking | 81.1 |
| 8 |  MiniMax M2.7 | 80.5 |
| 9 | GPT-5.1-Codex | 79.6 |
| 10 | MiniMax M2.5 | 77.4 |
| 11 |  MiMo-V2-Pro | 77.0 |
| 12 | GPT-5.1-Codex-Mini | 76.3 |
| 13 | GLM 4.7 | 76.0 |
| 14 | GPT-5 Mini | 74.4 |
| 15 | Qwen3 Next 80B A3B Thinking | 74.3 |
| 16 |  Gemma 4 31B | 73.9 |
| 17 | Qwen3 235B A22B Thinking 2507 | 73.4 |
| 18 | GLM 5V Turbo | 70.4 |
| 19 | Qwen3 Next 80B A3B Instruct | 70.2 |
| 20 | gpt-oss-120b | 68.9 |
| 21 | Qwen3 235B A22B Instruct 2507 | 68.0 |
| 22 | GPT-5 Nano | 64.7 |
| 23 |  DeepSeek V3.2 Exp | 64.4 |
| 24 | DeepSeek V3.2 | 64.0 |
| 25 | GLM 4.6V | 62.5 |
| 26 |  Devstral 2 2512 | 52.5 |
| 27 | GPT-5.4 Mini | 37.0 |
| 28 |  Nemotron 3 Super | 36.4 |
| 29 | GPT-5.4 Nano | 36.0 |
Frequently asked
Pulled from the LiveBench · Mathematics dataset · updated daily
What does LiveBench · Mathematics measure?
LiveBench · Mathematics is a knowledge benchmark in the BenchGecko catalog. 29 AI models have been tested on it. Scores range from 36.0 to 88.8 out of 100.
Which model leads on LiveBench · Mathematics?
GPT-5.2-Codex from OpenAI leads LiveBench · Mathematics with a score of 88.8. The median score across 29 tested models is 74.3.
Is LiveBench · Mathematics saturated?
No · the top score is 88.8 out of 100 (89%). There is still meaningful room for improvement on LiveBench · Mathematics.
Does LiveBench · Mathematics predict performance on other benchmarks?
Yes · LiveBench · Mathematics scores correlate 0.91 with LiveBench · Overall across 29 shared models. Models that do well on LiveBench · Mathematics tend to do well on LiveBench · Overall.
How often is LiveBench · Mathematics data refreshed?
BenchGecko pulls updates daily. New model scores on LiveBench · Mathematics appear as soon as they are published by Epoch AI or the model provider.
Top on LiveBench · Mathematics
GPT-5.2-Codex · 88.8GLM 5.1 · 84.9Qwen3.6 Plus · 83.7GPT-5.1-Codex-Max · 83.7GLM 5 · 83.5More knowledge benchmarks
Same category · related evaluations