GPQA Leaderboard
Compare models on GPQA benchmarks.
Last updated: May 22, 2025
Rank | Model | Score | Organization | License | Updated At |
---|---|---|---|---|---|
1 | google-gemini-2.5-pro-previewGoogle | 84 | Proprietary | May 18, 2025 | |
2 | o3-2025-04-16OpenAI | 83.3% | OpenAI | Proprietary | May 22, 2025 |
3 | o3OpenAI | 83.3% | OpenAI | Proprietary | May 22, 2025 |
4 | o4-miniOpenAI | 81.4% | OpenAI | Proprietary | May 22, 2025 |
5 | gpt-4o-miniOpenAI | 79.7% | OpenAI | Proprietary | May 22, 2025 |
6 | o3-miniOpenAI | 79.7% | OpenAI | Proprietary | May 22, 2025 |
7 | google-gemini-2.5-flash-previewGoogle | 78.3 | Proprietary | May 18, 2025 | |
8 | o1OpenAI | 75.7% | OpenAI | Proprietary | May 22, 2025 |
9 | claude-3-7-sonnet-20250219Anthropic | 68% | Anthropic | Proprietary | May 22, 2025 |
10 | claude-3-7-sonnet-latestAnthropic | 68% | Anthropic | Proprietary | May 22, 2025 |
11 | anthropic-claude-3-7-sonnetAnthropic | 68% | Anthropic | Proprietary | May 22, 2025 |
12 | gpt-4.1OpenAI | 66.3% | OpenAI | Proprietary | May 22, 2025 |
13 | anthropic-claude-3-5-sonnetAnthropic | 65% | Anthropic | Proprietary | May 22, 2025 |
14 | claude-3-5-sonnet-20241022Anthropic | 65 % | Anthropic | Proprietary | May 22, 2025 |
15 | claude-3-5-sonnet-20240620Anthropic | 65% | Anthropic | Proprietary | May 22, 2025 |
16 | gpt-4.1-miniOpenAI | 65% | OpenAI | Proprietary | May 22, 2025 |
17 | claude-3-5-sonnet-latestAnthropic | 65% | Anthropic | Proprietary | May 22, 2025 |
18 | google-gemini-2.0-flashGoogle | 62.1 | Proprietary | May 18, 2025 | |
19 | google-gemini-2.0-flash-liveGoogle | 62.1% | Proprietary | May 18, 2025 | |
20 | o1-miniOpenAI | 60% | OpenAI | Proprietary | May 22, 2025 |
21 | gpt-4o-2024-05-13OpenAI | 56.1% | OpenAI | Proprietary | May 22, 2025 |
22 | gpt-4o-2024-11-20OpenAI | 56.1% | OpenAI | Proprietary | May 22, 2025 |
23 | chatgpt-4oOpenAI | 56.1% | OpenAI | Proprietary | May 22, 2025 |
24 | gpt-4oOpenAI | 56.1% | OpenAI | Proprietary | May 22, 2025 |
25 | gpt-4o-2024-08-06OpenAI | 56.1% | OpenAI | Proprietary | May 22, 2025 |
26 | gpt-4.1-nanoOpenAI | 50.3 | OpenAI | Proprietary | May 22, 2025 |
27 | claude-3-5-haiku-20241022Anthropic | 41.6% | Anthropic | Proprietary | May 22, 2025 |
28 | anthropic-claude-3-5-haikuAnthropic | 41.6% | Anthropic | Proprietary | May 22, 2025 |
29 | claude-3-5-haiku-latestAnthropic | 41.6% | Anthropic | Proprietary | May 22, 2025 |