← All comparisons
Jamba 1.6 Large vs Qwen3 VL 32B (Reasoning)
AI21 Labs vs Alibaba — side-by-side benchmark comparison
| Jamba 1.6 Large | Qwen3 VL 32B (Reasoning) | |
|---|---|---|
| Intelligence Index | 10.6 | 24.7 |
| Coding Index | — | 14.5 |
| Math Index | — | 84.7 |
| Output speed (tok/s) | 54.9 | 96.3 |
| Blended price ($/1M) | $3.50 | $2.63 |
| Time to first token (s) | 0.87s | 1.12s |
| aime | 4.7% | — |
| aime 25 | — | 84.7% |
| artificial analysis coding index | — | 14.50 |
| artificial analysis intelligence index | 10.60 | 24.70 |
| artificial analysis math index | — | 84.70 |
| gpqa | 38.7% | 73.3% |
| hle | 4.0% | 9.6% |
| ifbench | — | 59.4% |
| lcr | — | 55.3% |
| livecodebench | 17.2% | 73.8% |
| math 500 | 58.0% | — |
| mmlu pro | 56.5% | 81.8% |
| scicode | 18.4% | 28.5% |
| tau2 | — | 45.6% |
| terminalbench hard | — | 7.6% |
Benchmark data from Artificial Analysis.