← All comparisons
GPT-3.5 Turbo vs Qwen3.5 27B (Non-reasoning)
OpenAI vs Alibaba — side-by-side benchmark comparison
| GPT-3.5 Turbo | Qwen3.5 27B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 9.0 | 37.2 |
| Coding Index | 10.7 | 33.4 |
| Math Index | — | — |
| Output speed (tok/s) | 116.9 | 95.3 |
| Blended price ($/1M) | $0.75 | $0.88 |
| Time to first token (s) | 0.56s | 1.40s |
| aime | — | — |
| aime 25 | — | — |
| artificial analysis coding index | 10.70 | 33.40 |
| artificial analysis intelligence index | 9.00 | 37.20 |
| artificial analysis math index | — | — |
| gpqa | 29.7% | 84.2% |
| hle | — | 13.2% |
| ifbench | — | 46.9% |
| lcr | — | 55.7% |
| livecodebench | — | — |
| math 500 | 44.1% | — |
| mmlu pro | 46.2% | — |
| scicode | — | 36.7% |
| tau2 | — | 87.1% |
| terminalbench hard | — | 31.8% |
Benchmark data from Artificial Analysis.