OverpayingForAIPricing desk

Lesson 3 of 8 · 9 min read · Beginner

Mistral pricing and plans: Le Chat, API and self-hosting

Le Chat Free, paid Le Chat tiers, API pay-as-you-go and running the open weights yourself. What each option costs, what it is worth, and which one fits you.

In this lesson you will

  • Compare Le Chat Free, paid Le Chat, API pay-as-you-go and self-hosting on cost and control
  • Estimate an API bill from tokens per task and tasks per month
  • Decide when a subscription is the cheaper option and when it is not

Mistral sells the same models through four doors, and the right door depends on how much you use and how much you care about control. This lesson keeps the numbers honest: API prices come from our live catalogue, and paid Le Chat prices are left to the official pricing page because they change and we do not track them.

Four ways to pay for MistralLe Chat Free
$0
per month
Chat with current Mistral models
Web search and file upload
Usage limits apply
Good enough to learn on
Le Chat paid tiers
See official page
per month
Higher limits and priority
Extra features vary by tier
Team and enterprise options
Check mistral.ai/pricing
API pay-as-you-go
Per 1M tokens
no monthly fee
Small 4: $0.15 in / $0.60 out
Large 3: $0.50 in / $1.50 out
Batch inference at a discount
Pin dated model ids
Self-host open weights
$0 per token
your hardware
Apache 2.0 downloads
No data leaves your network
You pay for GPUs and ops
Best above steady volume
Figure 1.API prices are from our catalogue at the time of writing. Paid Le Chat tiers are deliberately not priced here; use the official page.

Le Chat Free

The free tier is the right starting point for everyone. You get chat with current models, web search, file upload and image tools, subject to daily limits. If you have never used Mistral, spend a week here before you pay for anything. Most people discover that their real usage is a handful of conversations a day, which the free tier covers.

Paid Le Chat tiers

Mistral offers paid individual and team tiers with higher limits and extra features. We do not list their prices because they are not in our pricing catalogue and they change; check mistral.ai/pricing for the current figures. The buying test is simple: pay only when you hit the free limits on most days or need a specific paid feature. Do not pay for headroom you never use.

API pay-as-you-go

The API has no monthly fee. You pay per million tokens, input and output priced separately. A typical chat turn is around 500 tokens in and 300 out. On Mistral Small 4 that is about $0.00026 per turn, so a thousand turns cost roughly 26 cents. On Mistral Large 3 the same thousand turns cost about 70 cents. The API is nearly always cheaper than a subscription for light use; the subscription only wins when you run long, heavy sessions every day.

Subscription vs API: two ways to pay for the same modelFlat monthly fee with usage limits, or metered per-token billing with no ceiling.Subscription (chat app)API (pay per token)Fixed price every monthBill tracks real usageUsage caps and rate limitsNo cap — budget alerts are on youBest for daily heavy chatBest for bursts, automation, appsIncludes the app, files, memoryYou bring your own interfaceIdle months are wasted moneyIdle months cost $0Break-even test: if your API bill for the same work would be under the subscription price, the subscription is the overpayment.
Figure 2.A subscription is a flat fee; the API is a metered fee. Below the crossover the API is cheaper, above it the subscription is. For most individuals with Mistral the crossover is high because the per-token prices are low.

Self-hosting the open weights

Because Mistral publishes so many models under Apache 2.0, the fourth option is to run them yourself. The per-token price is zero, but the total cost is not: you pay for GPU time, someone to operate it, and the quality gap if you can only fit a smaller model. This wins when you have steady, predictable volume, a strict data-residency rule, or hardware that is already idle. It loses at low volume, where a rented GPU sitting mostly unused costs more than the API would.

Which door fits which user
You areStart withMove to
Curious, a few chats a dayLe Chat FreeA paid tier only if limits bite daily
Writer or analyst, many hours a dayLe Chat Free for a weekPaid Le Chat; compare on the official page
Developer building a featureAPI with mistral-small-latestPinned dated id, batch for bulk
EU team with residency rulesAPI on EU hostingSelf-host Small or Ministral at scale

Knowledge check

You send about 40 short prompts a day through Mistral Small 4 via the API. Roughly what is your monthly bill?

Lesson FAQ

How much does Le Chat Pro cost?

Paid Le Chat prices are not in our pricing catalogue, so we do not quote them. See the official page at mistral.ai/pricing for current individual and team tiers.

Does the Mistral API have a free tier?

Mistral has offered a free experimental API tier with strict rate limits at various times. Check console.mistral.ai for what is currently available; do not build production on it.

Is self-hosting Mistral cheaper than the API?

Only at steady, high volume or when you already own idle GPUs. At low volume the API is cheaper because you are not paying for hardware to sit idle.

Compare, calculate, decide — for Mistral (Le Chat)

If our calculators helped you cut down on hidden AI wallet leaks, consider buying us a coffee. A tiny fraction of your savings keeps our pricing indexes updated daily.

Not sure which AI is cheapest for your use case? Find out in 30 seconds — no signup required.

AI cost intelligence

Stop overpaying for AI tools

Join the OverpayingForAI list for pricing updates, cheaper alternatives, and practical buying guidance.

Now tracking 50+ AI tools, models, platforms, subscriptions, coding tools, and automation products.

We use your email only for OverpayingForAI updates. Unsubscribe anytime.