← All comparisons
Claude 3.5 Sonnet (Oct '24) vs GLM-4.5-Air
Anthropic vs Z AI — side-by-side benchmark comparison
| Claude 3.5 Sonnet (Oct '24) | GLM-4.5-Air | |
|---|---|---|
| Intelligence Index | 15.9 | 23.2 |
| Coding Index | 30.2 | 23.8 |
| Math Index | — | 80.7 |
| Output speed (tok/s) | 0.0 | 86.9 |
| Blended price ($/1M) | $6.56 | $0.37 |
| Time to first token (s) | 0.00s | 1.72s |
| aime | 15.7% | 67.3% |
| aime 25 | — | 80.7% |
| artificial analysis coding index | 30.20 | 23.80 |
| artificial analysis intelligence index | 15.90 | 23.20 |
| artificial analysis math index | — | 80.70 |
| gpqa | 59.9% | 73.3% |
| hle | 3.9% | 6.8% |
| ifbench | — | 37.6% |
| lcr | — | 43.7% |
| livecodebench | 38.1% | 68.4% |
| math 500 | 77.1% | 96.5% |
| mmlu pro | 77.2% | 81.5% |
| scicode | 36.6% | 30.6% |
| tau2 | — | 46.5% |
| terminalbench hard | — | 20.5% |
Benchmark data from Artificial Analysis.