DeepSeek V4 Pro vs GLM-5 Code
DeepSeek against Z.ai (GLM), costed on the same three workloads. Neither is cheaper everywhere — which one wins depends on how much you read versus write.
Monthly bill, side by side
| Workload | DeepSeek V4 Pro | GLM-5 Code | Difference |
|---|---|---|---|
| Support chatbot 2,500 six-turn conversations a day |
$4,247 | $4,428 | $181 (+4%) |
| Document summariser 1,000 single-shot summaries a day |
$578 | $559 | $18.96 (+3%) |
| Coding agent 200 twelve-turn sessions a day |
$3,231 | $3,190 | $41.76 (+1%) |
No prompt caching and no batch discount on either side, so the comparison is like for like. Both figures assume 30 days a month.
The rate cards
| DeepSeek V4 Pro | GLM-5 Code | |
|---|---|---|
| Input, per 1M tokens | $1.32 | $1.2 |
| Output, per 1M tokens | $3.96 | $5 |
| Cached input | $0.044 | $0.3 |
| Context window | 1M | 200K |
| Maximum output | 384K | 128K |
| Reasoning model | yes | yes |
| Vision | no | no |
Run this comparison on your own numbers
Full detail: DeepSeek V4 Pro pricing · GLM-5 Code pricing