← Benchmarks
Best AI Models for Reasoning Benchmark
Ranking of AI models by Reasoning benchmark score.
| Rank | Model | Provider | Score | Dataset | Version | Date |
|---|---|---|---|---|---|---|
| 1 | o3 | OpenAI | 96 | internal-demo | v1 | 2026-07-01 |
| 2 | Claude Opus 4 | Anthropic | 95 | internal-demo | v1 | 2026-07-01 |
| 3 | DeepSeek Reasoner | DeepSeek | 93 | internal-demo | v1 | 2026-07-01 |
| 4 | Gemini 2.5 Pro | 93 | internal-demo | v1 | 2026-07-01 | |
| 5 | Claude Sonnet 4 | Anthropic | 92 | internal-demo | v1 | 2026-07-01 |
| 6 | GPT-4.1 | OpenAI | 91 | internal-demo | v1 | 2026-07-01 |
| 7 | GPT-4o | OpenAI | 90 | internal-demo | v1 | 2026-07-01 |
| 8 | Gemini 2.5 Flash | 86 | internal-demo | v1 | 2026-07-01 | |
| 9 | DeepSeek Chat | DeepSeek | 85 | internal-demo | v1 | 2026-07-01 |