Z.ai: GLM 5.2 wins on more metrics (8 of 9), but the right pick depends on what you optimize for — see the breakdown below.
| Metric | Anthropic: Claude Sonnet 4.6 | Z.ai: GLM 5.2 |
|---|---|---|
| Intelligence Index | — | 51.1✓ |
| Coding Index | 43.0 | 68.8✓ |
| GPQA Diamond | 80% | 90%✓ |
| Design Arena Elo | 1326 | 1369✓ |
| Speed (tokens/sec) | 54 | 58✓ |
| Latency | 1.0s✓ | 1.2s |
| Input price /M | $3.00 | $1.20✓ |
| Output price /M | $15.00 | $4.10✓ |
| Context window | 1M | 1.0M✓ |
| Capabilities | ReasoningToolsJSONVision | ReasoningToolsJSON |