← All comparisons
Claude 4.1 Opus (Non-reasoning) vs GLM-4.6V (Reasoning)
Anthropic vs Z AI — side-by-side benchmark comparison
| Claude 4.1 Opus (Non-reasoning) | GLM-4.6V (Reasoning) | |
|---|---|---|
| Intelligence Index | 36.0 | 23.4 |
| Coding Index | — | 19.7 |
| Math Index | — | 85.3 |
| Output speed (tok/s) | 44.7 | 42.1 |
| Blended price ($/1M) | $32.81 | $0.45 |
| Time to first token (s) | 1.63s | 1.47s |
| aime | — | — |
| aime 25 | — | 85.3% |
| artificial analysis coding index | — | 19.70 |
| artificial analysis intelligence index | 36.00 | 23.40 |
| artificial analysis math index | — | 85.30 |
| gpqa | — | 71.9% |
| hle | — | 8.9% |
| ifbench | — | 30.1% |
| lcr | — | 40.3% |
| livecodebench | — | 16.0% |
| math 500 | — | — |
| mmlu pro | — | 79.9% |
| scicode | — | 30.4% |
| tau2 | — | 31.6% |
| terminalbench hard | — | 14.4% |
Benchmark data from Artificial Analysis.