← All comparisons
Motif-2-12.7B-Reasoning vs Qwen3.5 4B (Non-reasoning)
Motif Technologies vs Alibaba — side-by-side benchmark comparison
| Motif-2-12.7B-Reasoning | Qwen3.5 4B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 19.1 | 22.6 |
| Coding Index | 11.9 | 13.7 |
| Math Index | 80.3 | — |
| Output speed (tok/s) | 0.0 | 210.0 |
| Blended price ($/1M) | $0.00 | $0.06 |
| Time to first token (s) | 0.00s | 0.23s |
| aime | — | — |
| aime 25 | 80.3% | — |
| artificial analysis coding index | 11.90 | 13.70 |
| artificial analysis intelligence index | 19.10 | 22.60 |
| artificial analysis math index | 80.30 | — |
| gpqa | 69.5% | 71.2% |
| hle | 8.2% | 7.5% |
| ifbench | 57.0% | 33.3% |
| lcr | 13.0% | 28.3% |
| livecodebench | 65.1% | — |
| math 500 | — | — |
| mmlu pro | 79.6% | — |
| scicode | 28.2% | 18.3% |
| tau2 | 46.5% | 87.7% |
| terminalbench hard | 3.8% | 11.4% |
Benchmark data from Artificial Analysis.