← All comparisons
Tiny Aya Global vs Qwen3.5 27B (Non-reasoning)
Cohere vs Alibaba — side-by-side benchmark comparison
| Tiny Aya Global | Qwen3.5 27B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 4.7 | 37.2 |
| Coding Index | 1.2 | 33.4 |
| Math Index | — | — |
| Output speed (tok/s) | 124.0 | 95.3 |
| Blended price ($/1M) | $0.00 | $0.88 |
| Time to first token (s) | 0.32s | 1.40s |
| aime | — | — |
| aime 25 | — | — |
| artificial analysis coding index | 1.20 | 33.40 |
| artificial analysis intelligence index | 4.70 | 37.20 |
| artificial analysis math index | — | — |
| gpqa | 30.5% | 84.2% |
| hle | 5.2% | 13.2% |
| ifbench | 20.1% | 46.9% |
| lcr | 0.0% | 55.7% |
| livecodebench | — | — |
| math 500 | — | — |
| mmlu pro | — | — |
| scicode | 3.6% | 36.7% |
| tau2 | 0.0% | 87.1% |
| terminalbench hard | 0.0% | 31.8% |
Benchmark data from Artificial Analysis.