Z.ai: GLM 4.7 wins on more metrics (6 of 9), but the right pick depends on what you optimize for — see the breakdown below.
| Metric | DeepSeek: DeepSeek V3.2 | Z.ai: GLM 4.7 |
|---|---|---|
| Intelligence Index | 41.7 | 42.1✓ |
| Coding Index | 36.7✓ | 36.3 |
| GPQA Diamond | 84% | 86%✓ |
| Design Arena Elo | 1221 | 1274✓ |
| Speed (tokens/sec) | 41 | 179✓ |
| Latency | 458ms | 328ms✓ |
| Input price /M | $0.229✓ | $0.400 |
| Output price /M | $0.343✓ | $1.75 |
| Context window | 131K | 203K✓ |
| Capabilities | ReasoningToolsJSON | ReasoningToolsJSON |