Head to head
DeepSeek: DeepSeek V4 Pro 0423 vs Z.ai: GLM 5
Live prices, context windows, capabilities, and measured quality — updated continuously.
Side by side
| Metric | DeepSeek: DeepSeek V4 Pro 0423 Deepseek | Z.ai: GLM 5 Z Ai |
|---|---|---|
| Input $/M tokens | $0.660 | $1.00 |
| Output $/M tokens | $1.98 | $3.20 |
| Context window | 1.0M | 205K |
| Max output | 384K | 128K |
| WM Score | 86 | 81 |
| Benchmark quality | 82.4 | 76.6 |
| Hosts serving it | 18 | 11 |
| Throughput (tok/s) | 69 | 101 |
| Latency (ms) | 397 | 693 |
| Vision | ||
| Tools | ||
| Reasoning | ||
| JSON mode |
Which should you pick?
DeepSeek: DeepSeek V4 Pro 0423 costs less per million input tokens — about 1.5× cheaper than Z.ai: GLM 5. DeepSeek: DeepSeek V4 Pro 0423 scores higher on measured quality. DeepSeek: DeepSeek V4 Pro 0423 takes a larger context window, which matters for long documents or big codebases. Prices here are live list prices; the same model can cost less on a different host, so check the model page for the cheapest offering.