← All comparisons
Claude 4.1 Opus (Non-reasoning) vs GLM-4.7 (Reasoning)
Anthropic vs Z AI — side-by-side benchmark comparison
| Claude 4.1 Opus (Non-reasoning) | GLM-4.7 (Reasoning) | |
|---|---|---|
| Intelligence Index | 36.0 | 42.1 |
| Coding Index | — | 36.3 |
| Math Index | — | 95.0 |
| Output speed (tok/s) | 44.7 | 79.6 |
| Blended price ($/1M) | $32.81 | $1.00 |
| Time to first token (s) | 1.63s | 0.73s |
| aime | — | — |
| aime 25 | — | 95.0% |
| artificial analysis coding index | — | 36.30 |
| artificial analysis intelligence index | 36.00 | 42.10 |
| artificial analysis math index | — | 95.00 |
| gpqa | — | 85.9% |
| hle | — | 25.1% |
| ifbench | — | 67.9% |
| lcr | — | 64.0% |
| livecodebench | — | 89.4% |
| math 500 | — | — |
| mmlu pro | — | 85.6% |
| scicode | — | 45.1% |
| tau2 | — | 95.9% |
| terminalbench hard | — | 31.8% |
Benchmark data from Artificial Analysis.