← All comparisons
Grok 4 vs Qwen3 30B A3B 2507 (Reasoning)
xAI vs Alibaba — side-by-side benchmark comparison
| Grok 4 | Qwen3 30B A3B 2507 (Reasoning) | |
|---|---|---|
| Intelligence Index | 41.5 | 22.4 |
| Coding Index | 40.5 | 14.6 |
| Math Index | 92.7 | 56.3 |
| Output speed (tok/s) | 0.0 | 155.3 |
| Blended price ($/1M) | $11.00 | $0.67 |
| Time to first token (s) | 0.00s | 1.02s |
| aime | 94.3% | 90.7% |
| aime 25 | 92.7% | 56.3% |
| artificial analysis coding index | 40.50 | 14.60 |
| artificial analysis intelligence index | 41.50 | 22.40 |
| artificial analysis math index | 92.70 | 56.30 |
| gpqa | 87.7% | 70.7% |
| hle | 23.9% | 9.8% |
| ifbench | 53.7% | 50.7% |
| lcr | 68.0% | 59.0% |
| livecodebench | 81.9% | 70.7% |
| math 500 | 99.0% | 97.6% |
| mmlu pro | 86.6% | 80.5% |
| scicode | 45.7% | 33.3% |
| tau2 | 74.9% | 28.1% |
| terminalbench hard | 37.9% | 5.3% |
Benchmark data from Artificial Analysis.