← All comparisons
GPT-5.4 (Non-reasoning) vs DeepSeek R1 Distill Qwen 32B
OpenAI vs DeepSeek — side-by-side benchmark comparison
| GPT-5.4 (Non-reasoning) | DeepSeek R1 Distill Qwen 32B | |
|---|---|---|
| Intelligence Index | 35.4 | 17.2 |
| Coding Index | 41.0 | — |
| Math Index | — | 63.0 |
| Output speed (tok/s) | 70.9 | 0.0 |
| Blended price ($/1M) | $5.63 | $0.00 |
| Time to first token (s) | 0.60s | 0.00s |
| aime | — | 68.7% |
| aime 25 | — | 63.0% |
| artificial analysis coding index | 41.00 | — |
| artificial analysis intelligence index | 35.40 | 17.20 |
| artificial analysis math index | — | 63.00 |
| gpqa | 74.8% | 61.5% |
| hle | 10.6% | 5.5% |
| ifbench | 48.4% | 22.9% |
| lcr | 47.3% | 9.7% |
| livecodebench | — | 27.0% |
| math 500 | — | 94.1% |
| mmlu pro | — | 73.9% |
| scicode | 47.1% | 37.6% |
| tau2 | 35.1% | — |
| terminalbench hard | 37.9% | — |
Benchmark data from Artificial Analysis.