Mistral models through odnoga
European, small and cheap. Mistral is the value end of the gateway for tool-calling work that does not need a frontier model — and a useful answer when a customer asks who processes their data and where.
The cheap tier that still calls tools
Tool calling is what separates a model you can build a product on from a model that can only chat, and it is available across this family at prices near the bottom of the gateway. For extraction, routing, form-filling and other structured work at volume, that combination is hard to beat on cost per useful call.
No cached-input rate
Unlike most of the gateway, these models publish no cached-input price, so a long repeated prefix costs full price every time. If your prompt carries a large fixed preamble, compare the true per-call cost against a model that does cache before assuming the cheaper list price wins — the caching discount elsewhere is large enough to reverse the comparison.
Vendor cost is the list price. Your price is that plus your plan margin — the same figures that appear on your invoice.
Every model here, cheapest first
| Model | Vendor | Context | In / 1M | Out / 1M | Cached in | Your price in | |
|---|---|---|---|---|---|---|---|
| Ministral 3B ministral-3b-latest | Mistral AI | 131K | $0.1 | $0.1 | — | $0.108 | Details |
| Ministral 8B ministral-8b-latest | Mistral AI | 262K | $0.15 | $0.15 | — | $0.161 | Details |
| Mistral Small 3 mistral-small-latest | Mistral AI | 262K | $0.15 | $0.6 | — | $0.161 | Details |
| Codestral codestral-latest | Mistral AI | 256K | $0.3 | $0.9 | — | $0.323 | Details |
Your price column is the vendor cost +7%.
Narrow it differently
Models that can decide to call a function you defined and use the result.
The widest range on the gateway, and the widest price range with it — from the cheapest token on odnoga to the most expensive.
The best price per token of context on the gateway.
Or read how the gateway picks between them: routing and fallback, and what it costs: plans and margins.