← All comparisons
GPT-5.4 mini (xhigh) vs Qwen3 VL 235B A22B (Reasoning)
OpenAI vs Alibaba — side-by-side benchmark comparison
| GPT-5.4 mini (xhigh) | Qwen3 VL 235B A22B (Reasoning) | |
|---|---|---|
| Intelligence Index | 48.9 | 27.6 |
| Coding Index | 51.5 | 20.9 |
| Math Index | — | 88.3 |
| Output speed (tok/s) | 182.8 | 35.6 |
| Blended price ($/1M) | $1.69 | $2.17 |
| Time to first token (s) | 4.25s | 5.14s |
| aime | — | — |
| aime 25 | — | 88.3% |
| artificial analysis coding index | 51.50 | 20.90 |
| artificial analysis intelligence index | 48.90 | 27.60 |
| artificial analysis math index | — | 88.30 |
| gpqa | 87.5% | 77.2% |
| hle | 26.6% | 10.1% |
| ifbench | 73.3% | 56.5% |
| lcr | 69.3% | 58.7% |
| livecodebench | — | 64.6% |
| math 500 | — | — |
| mmlu pro | — | 83.6% |
| scicode | 49.9% | 39.9% |
| tau2 | 83.3% | 54.1% |
| terminalbench hard | 52.3% | 11.4% |
Benchmark data from Artificial Analysis.