LiveBench · Overall
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 · Overall
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 | 74.3 |
| 2 | GPT-5.1-Codex-Max | 72.0 |
| 3 |  Qwen3.6 Plus | 70.8 |
| 4 |  GLM 5.1 | 70.2 |
| 5 | GLM 5 | 68.8 |
| 6 | GPT-5.1-Codex | 68.6 |
| 7 |  MiniMax M2.7 | 63.5 |
| 8 |  Gemma 4 31B | 61.6 |
| 9 |  Kimi K2 Thinking | 61.6 |
| 10 | GPT-5 Mini | 61.0 |
| 11 | GPT-5.1-Codex-Mini | 60.4 |
| 12 | MiniMax M2.5 | 60.1 |
| 13 |  MiMo-V2-Pro | 58.1 |
| 14 | GLM 4.7 | 58.1 |
| 15 | GLM 4.6 | 55.2 |
| 16 | Qwen3 235B A22B Thinking 2507 | 53.0 |
| 17 |  DeepSeek V3.2 | 51.8 |
| 18 | Qwen3 Next 80B A3B Thinking | 50.4 |
| 19 | DeepSeek V3.2 Exp | 49.9 |
| 20 | GLM 5V Turbo | 49.6 |
| 21 | Qwen3 235B A22B Instruct 2507 | 48.8 |
| 22 | GPT-5 Nano | 48.6 |
| 23 | Qwen3 Next 80B A3B Instruct | 48.4 |
| 24 | gpt-oss-120b | 46.1 |
| 25 |  Devstral 2 2512 | 41.2 |
| 26 | GLM 4.6V | 40.1 |
| 27 | GPT-5.4 Mini | 37.0 |
| 28 |  Nemotron 3 Super | 32.5 |
| 29 | GPT-5.4 Nano | 32.4 |
Frequently asked
Pulled from the LiveBench · Overall dataset · updated daily
What does LiveBench · Overall measure?
LiveBench · Overall is a knowledge benchmark in the BenchGecko catalog. 29 AI models have been tested on it. Scores range from 32.4 to 74.3 out of 100.
Which model leads on LiveBench · Overall?
GPT-5.2-Codex from OpenAI leads LiveBench · Overall with a score of 74.3. The median score across 29 tested models is 55.2.
Is LiveBench · Overall saturated?
No · the top score is 74.3 out of 100 (74%). There is still meaningful room for improvement on LiveBench · Overall.
Does LiveBench · Overall predict performance on other benchmarks?
Yes · LiveBench · Overall scores correlate 0.92 with LiveBench · Reasoning across 29 shared models. Models that do well on LiveBench · Overall tend to do well on LiveBench · Reasoning.
How often is LiveBench · Overall data refreshed?
BenchGecko pulls updates daily. New model scores on LiveBench · Overall appear as soon as they are published by Epoch AI or the model provider.
Top on LiveBench · Overall
GPT-5.2-Codex · 74.3GPT-5.1-Codex-Max · 72.0Qwen3.6 Plus · 70.8GLM 5.1 · 70.2GLM 5 · 68.8More knowledge benchmarks
Same category · related evaluations