Command A vs GLM-5
Cohere against Z.ai (GLM), costed on the same three workloads. GLM-5 is cheaper on all three workloads.
Monthly bill, side by side
| Workload | Command A | GLM-5 | Difference |
|---|---|---|---|
| Support chatbot 2,500 six-turn conversations a day |
$9,056 | $3,299 | $5,758 (2.7×) |
| Document summariser 1,000 single-shot summaries a day |
$1,155 | $443 | $712 (2.6×) |
| Coding agent 200 twelve-turn sessions a day |
$6,570 | $2,484 | $4,086 (2.6×) |
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
| Command A | GLM-5 | |
|---|---|---|
| Input, per 1M tokens | $2.5 | $1 |
| Output, per 1M tokens | $10 | $3.2 |
| Cached input | not published | $0.2 |
| Context window | 256K | 200K |
| Maximum output | 8K | 128K |
| Reasoning model | no | yes |
| Vision | no | no |
Run this comparison on your own numbers
Full detail: Command A pricing · GLM-5 pricing