← All comparisons
GPT-4o (May '24) vs Gemini 2.0 Pro Experimental (Feb '25)
OpenAI vs Google — side-by-side benchmark comparison
| GPT-4o (May '24) | Gemini 2.0 Pro Experimental (Feb '25) | |
|---|---|---|
| Intelligence Index | 14.5 | 18.1 |
| Coding Index | 24.2 | 25.5 |
| Math Index | — | — |
| Output speed (tok/s) | 111.8 | 0.0 |
| Blended price ($/1M) | $7.50 | $0.00 |
| Time to first token (s) | 0.61s | 0.00s |
| aime | 11.0% | 36.0% |
| aime 25 | — | — |
| artificial analysis coding index | 24.20 | 25.50 |
| artificial analysis intelligence index | 14.50 | 18.10 |
| artificial analysis math index | — | — |
| gpqa | 52.6% | 62.2% |
| hle | 2.8% | 6.8% |
| ifbench | — | — |
| lcr | — | — |
| livecodebench | 33.4% | 34.7% |
| math 500 | 79.1% | 92.3% |
| mmlu pro | 74.0% | 80.5% |
| scicode | 30.9% | 31.3% |
| tau2 | — | — |
| terminalbench hard | — | — |
Benchmark data from Artificial Analysis.