pricing

start with GLM 5.3 Flash RR. it's our default for agents: $0.20 in / $0.08 cached / $0.80 out per 1M tokens, 1M context. you only pay for tokens you use.

every other model and price is on the models page. the GLM 5.3 Flash RR page has a curl example, and the quickstart gets you from key to first request.

monthly plans

each plan includes its price back as credit at standard rates, plus higher limits. plan credit is spent first and expires at the end of the period.

every plan covers every model, in chat and over the API. sign up to start one.

pay as you go

no plan needed. add credits by card or USDC, starting at $10. top-ups never expire.

zero data retention

zero data retention is an opt-in switch in settings. it's off until you turn it on. it's free on monthly plans. on pay as you go it adds 20% to every token price.

while it's on, we store no prompts or completions, and saving chats is off. usage metadata is still recorded for billing. what redrouter keeps has the full policy.

faq

how do i turn on zero data retention?

flip the switch in settings. it's off by default, and it applies to requests made after you turn it on.

does plan credit roll over?

no. plan credit expires at the end of the period. top-ups don't expire.

what happens when i change plans?

the change applies at renewal. there's no proration.

when does the cached input price apply?

only when a cache read is reported for the request. everything else bills at the standard input rate.

do you hold credit while a request runs?

yes. we reserve credit before a request starts and settle it when usage comes back. credits and billing explains holds.