25
runs, one time
Daily 5 · Parallel 1 agent · Card None
- Every hosted model
- Top-ups later, same account
One monthly number, paced by a daily cap. Your own key stays free forever.
25
runs, one time
Daily 5 · Parallel 1 agent · Card None
180
runs a month
Daily 8 · Parallel 1 agent · Resets Monthly
500
runs a month
Daily 20 · Parallel 3 agents · Resets Monthly
150,000 tokens of model work across the whole mission. You see the cost before a heavy mission starts, not after it ends.
The daily cap spreads 180 runs over about 23 days, so automation keeps running through the month instead of stopping on day nine.
60 runs for $6, prepaid, no expiry and outside the daily cap. Nothing ever auto-charges.
One run is up to 150,000 tokens of model work, counted across every agent, retry and tool call in a mission. A scheduled check usually spends well under one. A heavy coding mission can spend several, and it tells you before it starts.
The mission already running finishes. New missions are blocked by one card that shows your reset date and a top-up. Prepaid top-ups never expire, and nothing charges your card automatically.
Yes, and it stays free forever. Hosted runs exist for people who would rather not deal with provider keys, billing dashboards and rate limits.
Three pinned models: DeepSeek-V4.1-Flash-1017 as the default, with GLM-5.3-Flash and Qwen3.8-Flash as fallbacks. There is no model picker, because the app picks the right one and that is what keeps the price this low.
The monthly allowance resets on your plan date and does not accumulate. Prepaid top-ups are different: they sit in your account and never expire.
From the billing portal, in two clicks. You keep the plan until the end of the paid month, then drop to bring-your-own-key. No retention call, no form.
Charges are handled by Stripe. Read the Terms, Refund Policy and Privacy Policy.