← All comparisons
Qwen3.5 397B A17B (Reasoning) vs DeepSeek LLM 67B Chat (V1)
Alibaba vs DeepSeek — side-by-side benchmark comparison
| Qwen3.5 397B A17B (Reasoning) | DeepSeek LLM 67B Chat (V1) | |
|---|---|---|
| Intelligence Index | 45.0 | 8.4 |
| Coding Index | 41.3 | — |
| Math Index | — | — |
| Output speed (tok/s) | 52.1 | 0.0 |
| Blended price ($/1M) | $1.35 | $0.00 |
| Time to first token (s) | 1.81s | 0.00s |
| aime | — | — |
| aime 25 | — | — |
| artificial analysis coding index | 41.30 | — |
| artificial analysis intelligence index | 45.00 | 8.40 |
| artificial analysis math index | — | — |
| gpqa | 89.3% | — |
| hle | 27.3% | — |
| ifbench | 78.8% | — |
| lcr | 65.7% | — |
| livecodebench | — | — |
| math 500 | — | — |
| mmlu pro | — | — |
| scicode | 42.0% | — |
| tau2 | 95.6% | — |
| terminalbench hard | 40.9% | — |
Benchmark data from Artificial Analysis.