Field manual
Six chapters on plans, prices, timing and limits for Ollama Cloud. Ollama’s own pages and our live probes are the sources. Our own tests are marked as ours.
Verified 12 Oct 2026
Chapters
- Plans and quotaNew dollar credits against the old session and weekly windows, and how to tell which plan you are on.
- Cheaper modelsPer-token prices, cached input, the off-peak window, and what our per-call tests showed.
- Caching and tokensWhy stable prompt prefixes decide most of an agent bill, and how thinking models use max_tokens.
- Schedule around resetsMonthly, five-hour and weekly resets, off-peak windows for batch work, and concurrency limits.
- Agents and automationThe base URL, model names, 429 and 502 handling, fallback chains, and what the cloud API does not do.
- When cappedThe exact monthly limit message, what it means on each plan, and how to see it coming.
Check the real numbers
The manual explains the rules. The checker shows your own key’s windows, with estimates labelled EST. and exact dollars labelled EXACT.
Check your own usage