← All comparisons
LFM 40B vs Qwen3 VL 8B (Reasoning)
Liquid AI vs Alibaba — side-by-side benchmark comparison
| LFM 40B | Qwen3 VL 8B (Reasoning) | |
|---|---|---|
| Intelligence Index | 8.8 | 16.7 |
| Coding Index | — | 9.8 |
| Math Index | — | 30.7 |
| Output speed (tok/s) | 0.0 | 133.6 |
| Blended price ($/1M) | $0.00 | $0.66 |
| Time to first token (s) | 0.00s | 1.13s |
| aime | 2.3% | — |
| aime 25 | — | 30.7% |
| artificial analysis coding index | — | 9.80 |
| artificial analysis intelligence index | 8.80 | 16.70 |
| artificial analysis math index | — | 30.70 |
| gpqa | 32.7% | 57.9% |
| hle | 4.9% | 3.3% |
| ifbench | — | 39.9% |
| lcr | — | 31.0% |
| livecodebench | 9.6% | 35.3% |
| math 500 | 48.0% | — |
| mmlu pro | 42.5% | 74.9% |
| scicode | 7.1% | 21.9% |
| tau2 | — | 22.5% |
| terminalbench hard | — | 3.8% |
Benchmark data from Artificial Analysis.