Mistral through Squiid

What Mistral is

Mistral AI publishes both open-weight models and hosted frontier models, served from La Plateforme. Beyond chat completions the API includes embeddings, function calling, a fill-in-the-middle endpoint aimed at code completion, and Mistral OCR for turning documents into structured text. Being a European company with EU-hosted inference, it is the provider teams pick when data residency is part of the requirement rather than an afterthought. Codestral targets code completion specifically, and the open-weight releases mean you can prototype against the hosted API and self-host the same weights later if you need to.

Inside an AI-coded app Mistral fits the workloads where cost per token and jurisdiction matter: bulk classification, summarising support threads, embedding a document corpus before it goes into a vector store, or reading uploaded invoices with OCR. Its smaller models are cheap enough to call on every row of a table, which is where a per-token bill can surprise you if nothing is capping it. Mistral's function calling and JSON output behave like the other providers', so a router that already speaks one API can add it without a new abstraction layer.

Why connect it through Squiid

Squiid holds the Mistral key and meters every call against your prepaid balance, so the surprise is bounded by the credits you funded. Requests go to https://api.squiid.io/v1/mistral/... with Authorization: Bearer $SQUIID_API_KEY, and Mistral's client libraries work by pointing their base URL at the gateway (gateway convention). Usage appears per model in your dashboard next to your database and hosting, and settles on one invoice at Mistral's list price with no Squiid margin. OCR pages and embedding calls are itemised separately from chat tokens in the usage view.

• One account, one key, one bill, one dashboard. Mistral sits beside every other service your project uses, on the same invoice and the same credit balance.
• Key custody. Squiid holds the Mistral credentials. Your code carries SQUIID_API_KEY, so MISTRAL_API_KEY, MISTRAL_BASE_URL never land in a repository, a preview deployment or an agent’s context.
• Metering you can see. Every call is counted per service and per day, so you know what Mistral costs before the month ends.
• Pause on zero. When prepaid credits run out the gateway stops authorizing usage instead of running up a bill you did not fund.
• One rotation. Rotating the Squiid key cuts off every agent, worker and deployment at once, with no provider key to hunt down.

What you get

• Mistral chat, code and embedding models
• Mistral OCR for document extraction on the same key
• EU-hosted inference through the gateway
• Per-model spend metering and credit caps
• Mistral list pricing, no markup

How to connect Mistral

One variable replaces the provider credentials this service would normally need. Requests go to the gateway and the official SDKs work by pointing their base URL at it (gateway convention).

# .env
# replaces MISTRAL_API_KEY, MISTRAL_BASE_URL
SQUIID_API_KEY=sq_live_xxxxxxxxxxxxxxxxxxxxxxxx

# every request: https://api.squiid.io/v1/mistral/...
# header:       Authorization: Bearer $SQUIID_API_KEY

Pricing through Squiid

Mistral charges per input and output token by model, with separate rates for OCR pages and embeddings, at the provider's listed price. Squiid passes that through at par: one dollar of prepaid credit is one dollar of Mistral usage, with no markup and no per-request margin.

Mistral has no subscription attached on Squiid today. If a plan is required for a feature you need, it passes through at the provider’s listed price with no Squiid fee.

Squiid adds no markup to provider prices. There is a free plan at $0 a month with a 15% fee when you top up credits, Solo at $19.97 a month with a top-up fee that starts at 8% and falls to 3.5% as your spend grows, Team from $29.97 a seat a month with the same ladder one point lower, and Custom with a rate of 3% or less. Credits are prepaid, so Mistral usage stops when the balance reaches zero rather than producing an invoice you did not fund. Auto top-up with a cap keeps production running without handing anyone an open-ended card.

If you leave

Nothing to move: Mistral is usage-metered, so leaving means creating your own Mistral key and pointing the client at it. Squiid never owns your code, your schema or your domain, and the gateway speaks Mistral's own API, so the client change is a base URL. The handover page lists what happens for every provider in the catalogue.

Works with every coding agent

Mistral through Squiid is just an HTTP call with a bearer token, so every agentic coding tool can use it with the single key already in your environment:

• Claude Code
• Cursor
• ChatGPT
• Codex
• Grok
• Gemini CLI
• GitHub Copilot
• Windsurf
• Perplexity
• Devin
• Replit
• v0
• Lovable
• Cline

Vibe coding a feature on Mistral usually means asking for it in plain language and letting the agent write the integration. Because the key is Squiid’s rather than the provider’s, it can do that without ever holding a credential that could be leaked, logged or committed.