← All comparisons
Apriel-v1.6-15B-Thinker vs GPT-5.1 (Non-reasoning)
ServiceNow vs OpenAI — side-by-side benchmark comparison
| Apriel-v1.6-15B-Thinker | GPT-5.1 (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 27.6 | 27.4 |
| Coding Index | 22.0 | 27.3 |
| Math Index | 88.0 | 38.0 |
| Output speed (tok/s) | 0.0 | 129.5 |
| Blended price ($/1M) | $0.00 | $3.44 |
| Time to first token (s) | 0.00s | 0.72s |
| aime | — | — |
| aime 25 | 88.0% | 38.0% |
| artificial analysis coding index | 22.00 | 27.30 |
| artificial analysis intelligence index | 27.60 | 27.40 |
| artificial analysis math index | 88.00 | 38.00 |
| gpqa | 73.3% | 64.3% |
| hle | 9.8% | 5.2% |
| ifbench | 69.1% | 43.2% |
| lcr | 50.3% | 44.0% |
| livecodebench | 80.7% | 49.4% |
| math 500 | — | — |
| mmlu pro | 79.0% | 80.1% |
| scicode | 37.3% | 36.5% |
| tau2 | 69.3% | 46.5% |
| terminalbench hard | 14.4% | 22.7% |
Benchmark data from Artificial Analysis.