AI Leaderboards
Top models ranked by performance.
266 models
| License | |||||||||
|---|---|---|---|---|---|---|---|---|---|
1 |  Claude Opus 4.6(Anthropic) | 2,002 | 1,476 | 91.3% | 80.8% | — | — | — | Proprietary |
2 |  Gemini 3.1 Pro(Google) | 1,855 | 1,222 | 94.3% | 80.6% | — | — | — | Proprietary |
3 |  GPT-5.4(OpenAI) | 1,665 | 1,146 | 92.8% | — | — | — | — | Proprietary |
4 |  GLM-5(Zhipu AI) | 1,600 | 1,179 | — | 77.8% | — | — | — | Open Source |
5 | Claude Opus 4.5(Anthropic) | 1,582 | 1,345 | 87.0% | 80.9% | — | — | — | Proprietary |
6 | Gemini 3 Flash(Google) | 1,581 | 1,172 | 90.4% | 78.0% | — | — | — | Proprietary |
7 | Gemini 3 Pro(Google) | 1,579 | 1,045 | 91.9% | 76.2% | — | — | — | Proprietary |
8 | GPT-5.2(OpenAI) | 1,505 | 1,172 | 92.4% | 80.0% | — | — | — | Proprietary |
9 |  Kimi K2.5(Moonshot AI) | 1,448 | 988 | 87.6% | 76.8% | — | — | — | Open Source |
10 | Claude Sonnet 4.6(Anthropic) | 1,371 | 941 | 89.9% | 79.6% | — | — | — | Proprietary |
11 | GPT-5 High(OpenAI) | 1,301 | 1,037 | 87.3% | — | — | — | — | Proprietary |
12 |  Qwen3.5-397B-A17B(Alibaba Cloud / Qwen Team) | 1,214 | 1,067 | 88.4% | 76.4% | — | — | — | Open Source |
13 | GLM-4.6(Zhipu AI) | 1,143 | 1,079 | 81.0% | 68.0% | — | — | — | Open Source |
14 | Qwen3.5-122B-A10B(Alibaba Cloud / Qwen Team) | 1,136 | — | 86.6% | 72.0% | — | — | — | Open Source |
15 | GPT-5.2 Codex(OpenAI) | 1,133 | 802 | — | — | — | — | — | Proprietary |
16 | GPT-5.1(OpenAI) | 1,125 | 1,010 | 88.1% | 76.3% | — | — | — | Proprietary |
17 | GPT-5.1 High(OpenAI) | 1,117 | 1,132 | 88.1% | — | — | — | — | Proprietary |
18 | Claude Sonnet 4.5(Anthropic) | 1,103 | 1,294 | 83.4% | — | — | — | — | Proprietary |
19 | GPT-5 Medium(OpenAI) | 1,098 | 1,026 | 88.1% | — | — | — | — | Proprietary |
20 | GPT-5.3 Codex(OpenAI) | 1,089 | 650 | — | — | — | — | — | Proprietary |
1-20 of 266
New Models
Announced in the last 15 days