← All comparisons
Granite 4.0 H 1B vs GPT-5 nano (minimal)
IBM vs OpenAI — side-by-side benchmark comparison
| Granite 4.0 H 1B | GPT-5 nano (minimal) | |
|---|---|---|
| Intelligence Index | 8.0 | 13.8 |
| Coding Index | 2.7 | 14.2 |
| Math Index | 6.3 | 27.3 |
| Output speed (tok/s) | 0.0 | 175.6 |
| Blended price ($/1M) | $0.00 | $0.14 |
| Time to first token (s) | 0.00s | 0.67s |
| aime | — | — |
| aime 25 | 6.3% | 27.3% |
| artificial analysis coding index | 2.70 | 14.20 |
| artificial analysis intelligence index | 8.00 | 13.80 |
| artificial analysis math index | 6.30 | 27.30 |
| gpqa | 26.3% | 42.8% |
| hle | 5.0% | 4.1% |
| ifbench | 26.2% | 32.5% |
| lcr | 6.3% | 20.0% |
| livecodebench | 11.5% | 47.0% |
| math 500 | — | — |
| mmlu pro | 27.7% | 55.6% |
| scicode | 8.2% | 29.1% |
| tau2 | 19.6% | 25.7% |
| terminalbench hard | 0.0% | 6.8% |
Benchmark data from Artificial Analysis.