Models · Mistral AI

Mistral Large 4, run by Mistral AI in its EU region.

Mistral Large 4 is a model by Mistral AI that reads text and images: neither its weights nor its licence are published. On Evanesca, Mistral AI runs it in its own EU region: your prompts are processed in the EU by the company that made the model, and Evanesca keeps no copy of them — Mistral AI keeps prompts and answers for 30 days to monitor abuse.

Checked on 20 Oct 2026

Mistral Large 4

by Mistral AI

EU region of Mistral AI Checks 4/6

Input / M: €2.5133
Output / M: €7.7246
Quick question: €0.0038

Prices per million tokens, VAT included. Quick question: 300 tokens in, 400 out.

The model at a glance

Maker: Mistral AI
Licence: — no licence published yet
Size: 1T (49B active) parameters
Context: 512k tokens as the operator announces it
Type: Text + images
Tool calling: Yes measured by a real call
Reasoning: —
Run by: Mistral AI Mistral AI SAS, Paris — serves its own EU region, no US parent
Where it runs: EU region of Mistral AI
Checks: 4/6 · Licence ✕ · Provenance ✕ Passed: Security, GDPR, Benchmarks and Live test
Price: €2.5133 per million input tokens · €7.7246 per million output tokens VAT included; reasoning tokens count as output
Open weights: No kept by the maker: API access only
Model card: mistral.ai/news/mistral-large-4

Where your prompts go

Here is the route of a message sent to Mistral Large 4, end to end. No call to the model leaves Europe: not as a backup, not when servers are busy.

From your browser to EvanescaEvanesca passes your message straight on to Mistral AI. It stores neither the prompt nor the answer: both stay in memory only while the answer is written.
Mistral AI, EU regionMistral AI runs Mistral Large 4 in its own EU region, and does not publish which EU country serves a request. It is a French company with no US parent: out of the CLOUD Act's direct reach. Mistral AI keeps API inputs and outputs “for thirty (30) rolling days to monitor abuse” (Mistral's privacy policy).
Back to you, with its passportThe answer comes back with its passport: model, operator, country, and what was kept. Your history lives in your browser; about this message, Evanesca keeps only billing data — tokens and cost.

What it costs

€2.5133 per million input tokens and €7.7246 per million output tokens, VAT included. Three typical exchanges on Mistral Large 4:

COST OF AN EXCHANGE · VAT INCLUDED
Example	Tokens in	Tokens out	Cost
A quick question	300	400	€0.0038
A long document to summarise	15,000	800	€0.0439
A message late in a long conversation	40,000	2,000	€0.1160

€2 of credit covers about 520 quick questions on this model. Reasoning tokens count as output, so a model that thinks first can cost more than these examples. The exact cost of each answer shows under it. All prices.

Licence

Mistral AI has not published the weights of Mistral Large 4 yet: for now, the model is offered through an API only. No licence has been published for this model yet (official page).

Our licence check is not confirmed for this model: read the maker's terms at the source before you rely on it.

The official model card is the reference.

Similar models

Same maker first, then the same type.

EU region of Mistral AI

Mistral Small 4

Mistral AI · Text + images · checks 5/6

€0.2772 · €1.1088 / M tokens

Paris, France

Mistral Small 3.2

Mistral AI · Text + images · checks 5/6

€0.252 · €0.588 / M tokens

Paris, France

Pixtral 12B

Mistral AI · Vision · checks 5/6

€0.336 · €0.336 / M tokens

Germany (AKI.IO)

DeepSeek V4.1 Flash

DeepSeek · Text + images · checks 6/6

€1.008 · €3.024 / M tokens

Paris, France

Qwen3.8 27B

Qwen · Text + images · checks 6/6

€1.008 · €5.544 / M tokens

Switzerland (Infomaniak)

Apertus v1.5 70B

Swiss AI Initiative · Text + images · checks 4/6

€1.2936 · €4.62 / M tokens

Every model, with its operator and where it runs

Frequently asked questions

Where does Mistral Large 4 run on Evanesca?

In Mistral AI's EU region. Mistral AI, which made Mistral Large 4, runs it itself and does not publish which EU country serves a request. Mistral AI is a French company with no US parent, so your prompts stay out of the CLOUD Act's direct reach. No call to the model leaves Europe: not as a backup, not when servers are busy.

Does Evanesca keep what I send to Mistral Large 4?

No. The prompt and the answer stay in memory only while the answer is written, and your history lives in your browser; about your messages, Evanesca keeps only billing data — tokens and cost. Mistral AI keeps API inputs and outputs “for thirty (30) rolling days to monitor abuse” (Mistral's privacy policy).

How much does Mistral Large 4 cost?

€2.5133 per million input tokens and €7.7246 per million output tokens, VAT included. A quick question — 300 tokens in, 400 out — costs about €0.0038. There is no subscription: credit starts at €2.

Can Mistral Large 4 call tools?

Yes: a real call through Mistral AI returned proper tool calls (function calling).

Which licence is Mistral Large 4 under?

No licence has been published for this model yet (official page).

Which checks has Mistral Large 4 passed?

4 of 6: Security, GDPR, Benchmarks and Live test. Missing: Licence and Provenance — not checked yet, or checked without meeting the pass mark. Its passport shows the same stamps; how each check works.

Does Mistral AI see what I type?

Yes: Mistral AI, which made Mistral Large 4, runs it in its own EU region, so your prompts go to its servers. Mistral AI keeps API inputs and outputs “for thirty (30) rolling days to monitor abuse” (Mistral's privacy policy). And since Mistral Large 4 is a preview, its commercial terms (§ 4.3) let it use what you send to train its models, whatever the customer's objection.