← All comparisons
Jamba 1.7 Large vs Qwen3 VL 8B (Reasoning)
AI21 Labs vs Alibaba — side-by-side benchmark comparison
| Jamba 1.7 Large | Qwen3 VL 8B (Reasoning) | |
|---|---|---|
| Intelligence Index | 10.9 | 16.7 |
| Coding Index | 7.8 | 9.8 |
| Math Index | 2.3 | 30.7 |
| Output speed (tok/s) | 56.9 | 133.6 |
| Blended price ($/1M) | $3.50 | $0.66 |
| Time to first token (s) | 0.90s | 1.13s |
| aime | 5.7% | — |
| aime 25 | 2.3% | 30.7% |
| artificial analysis coding index | 7.80 | 9.80 |
| artificial analysis intelligence index | 10.90 | 16.70 |
| artificial analysis math index | 2.30 | 30.70 |
| gpqa | 39.0% | 57.9% |
| hle | 3.8% | 3.3% |
| ifbench | 35.2% | 39.9% |
| lcr | 17.3% | 31.0% |
| livecodebench | 18.1% | 35.3% |
| math 500 | 60.0% | — |
| mmlu pro | 57.7% | 74.9% |
| scicode | 18.8% | 21.9% |
| tau2 | 13.5% | 22.5% |
| terminalbench hard | 2.3% | 3.8% |
Benchmark data from Artificial Analysis.