← All comparisons
GPT-4o (Aug '24) vs DeepSeek-V2-Chat
OpenAI vs DeepSeek — side-by-side benchmark comparison
| GPT-4o (Aug '24) | DeepSeek-V2-Chat | |
|---|---|---|
| Intelligence Index | 18.6 | 9.1 |
| Coding Index | 16.6 | — |
| Math Index | — | — |
| Output speed (tok/s) | 117.5 | 0.0 |
| Blended price ($/1M) | $4.38 | $0.00 |
| Time to first token (s) | 0.60s | 0.00s |
| aime | 11.7% | — |
| aime 25 | — | — |
| artificial analysis coding index | 16.60 | — |
| artificial analysis intelligence index | 18.60 | 9.10 |
| artificial analysis math index | — | — |
| gpqa | 52.1% | — |
| hle | 2.9% | — |
| ifbench | 36.0% | — |
| lcr | 35.0% | — |
| livecodebench | 31.7% | — |
| math 500 | 79.5% | — |
| mmlu pro | — | — |
| scicode | 33.1% | — |
| tau2 | 28.9% | — |
| terminalbench hard | 8.3% | — |
Benchmark data from Artificial Analysis.