← All comparisons
GPT-4o (March 2025, chatgpt-4o-latest) vs DeepSeek R1 Distill Llama 70B
OpenAI vs DeepSeek — side-by-side benchmark comparison
| GPT-4o (March 2025, chatgpt-4o-latest) | DeepSeek R1 Distill Llama 70B | |
|---|---|---|
| Intelligence Index | 18.6 | 16.0 |
| Coding Index | — | 11.4 |
| Math Index | 25.7 | 53.7 |
| Output speed (tok/s) | 0.0 | 46.8 |
| Blended price ($/1M) | $0.00 | $0.79 |
| Time to first token (s) | 0.00s | 0.33s |
| aime | 32.7% | 67.0% |
| aime 25 | 25.7% | 53.7% |
| artificial analysis coding index | — | 11.40 |
| artificial analysis intelligence index | 18.60 | 16.00 |
| artificial analysis math index | 25.70 | 53.70 |
| gpqa | 65.5% | 40.2% |
| hle | 5.0% | 6.1% |
| ifbench | — | 27.6% |
| lcr | — | 11.0% |
| livecodebench | 42.5% | 26.6% |
| math 500 | 89.3% | 93.5% |
| mmlu pro | 80.3% | 79.5% |
| scicode | 36.6% | 31.3% |
| tau2 | — | 21.9% |
| terminalbench hard | — | 1.5% |
Benchmark data from Artificial Analysis.