What Pricing Plans and Options Does Mistral Offer?

Mistral publishes three separate pricing tracks rather than one unified plan: a general Plans page, a dedicated API pricing page, and a For enterprises page for enterprise deployments. Which one applies to you depends on how you intend to consume the models — through a chat/assistant product, through the API, or through a custom deployment inside your own infrastructure. Exact figures live on those pages and change over time, so treat the official pricing pages as the source of truth rather than any secondhand summary.

The three pricing tracks

The distinction matters because the three are not interchangeable. A "Plans" tier typically bundles access to a product surface, while API pricing is metered by usage, and enterprise deployments are scoped around infrastructure and customization requirements.

API pricing

If you are building on Mistral models, the API pricing page is the relevant entry point. API pricing is usage-based, so your cost scales with the volume and type of requests you send rather than a flat subscription. To estimate spend, you need to know which models you'll call and your expected request volume — both of which you can map against the published rates on that page.

Enterprise deployments

The enterprise track is separate from both the Plans and API pages. Mistral describes itself as helping organizations "build tailored AI systems," and its enterprise material emphasizes in-region inference, open models, and European infrastructure for sovereign AI. Enterprise deployments are therefore typically scoped to an organization's specific requirements — including where inference runs and how models are customized — rather than sold as a fixed public price. If your requirements include data residency, custom fine-tuning, or self-hosting, start with the enterprise page and expect a scoped conversation rather than an off-the-shelf number.

How to choose

• Just trying the product or buying seats for a team? Start with the Plans page.
• Integrating models into your own application? Use the API pricing page and estimate from your expected usage.
• Need sovereign, in-region, or customized deployment? Go to the enterprise deployments page.

What to verify before committing

Because Mistral's pricing pages are the authoritative source and are updated over time, confirm the following directly on the relevant page before you budget:

• The current rate or tier structure for the specific model or plan you intend to use.
• Whether usage is metered, seat-based, or scoped per deployment.
• Any conditions attached to enterprise or in-region options, since these are typically defined per engagement.

The site's own framing — "Frontier AI. In your hands." and a focus on tailored systems for enterprises — signals that Mistral positions flexible, deployment-specific arrangements as a core part of its offering, which is why no single price list covers every case.