← All comparisons
Granite 4.0 H 1B vs o3-mini (high)
IBM vs OpenAI — side-by-side benchmark comparison
| Granite 4.0 H 1B | o3-mini (high) | |
|---|---|---|
| Intelligence Index | 8.0 | 25.2 |
| Coding Index | 2.7 | 17.3 |
| Math Index | 6.3 | — |
| Output speed (tok/s) | 0.0 | 187.0 |
| Blended price ($/1M) | $0.00 | $1.93 |
| Time to first token (s) | 0.00s | 15.73s |
| aime | — | 86.0% |
| aime 25 | 6.3% | — |
| artificial analysis coding index | 2.70 | 17.30 |
| artificial analysis intelligence index | 8.00 | 25.20 |
| artificial analysis math index | 6.30 | — |
| gpqa | 26.3% | 77.3% |
| hle | 5.0% | 12.3% |
| ifbench | 26.2% | 67.1% |
| lcr | 6.3% | 39.3% |
| livecodebench | 11.5% | 73.4% |
| math 500 | — | 98.5% |
| mmlu pro | 27.7% | 80.2% |
| scicode | 8.2% | 39.8% |
| tau2 | 19.6% | 31.3% |
| terminalbench hard | 0.0% | 6.1% |
Benchmark data from Artificial Analysis.