Mistral Le Chat (Vibe) Review 2026: Pricing, Features and Verdict

What it is

Le Chat was Mistral AI's chat assistant. Mistral, a European AI lab, renamed it Vibe and merged its coding agent into the same product. The App Store listing now reads Vibe by Mistral (ex Le Chat). If you search for Le Chat, this is the product you will land on.

Vibe covers chat with web search, document creation (PDFs, presentations, reports), deep research across the web and your own documents, data and spreadsheet analysis, email drafting and task scheduling. Its code side writes, tests and refactors code in the CLI, an IDE or on the web. It connects to more than 100 tools including email, calendar, Slack, GitHub and Jira, with full MCP support.

Pricing explained

Plan	Price	Key limits and inclusions
Free	$0	Limited messages and web searches, basic image generation, limited coding, 100+ connectors
Pro	$14.99 per month excluding taxes	Up to 6x Free messages, up to 40x image generations, 15 GB libraries, all day coding, $15 per month API credits, chat and email support
Pro for students	$5.99 per month	Verified students at accredited institutions, new users, up to 12 months
Team	$24.99 per user per month, $50 minimum	Higher limits than Pro, 30 GB per user, domain verification, data export
Enterprise	Contact sales	Custom models, agents and workflows, audit logs, SAML SSO, white label, on premises or private cloud

Pro at $14.99 is the lowest price for a paid tier from a major lab other than Google AI Plus at $4.99 and ChatGPT Go at $8. Unlike those two, Vibe Pro includes a coding agent and API credits. The $15 in monthly API credits effectively refunds the subscription for developers who call Mistral's API anyway.

Team at $24.99 per user requires a $50 monthly minimum, so the smallest team pays for two seats. Mistral does not list annual billing on its pricing page.

The weak spot is model transparency. The pricing page does not name the model behind each plan; Mistral's docs recommend Mistral Medium for most tasks. On the API, Mistral Large lists at $0.50 per 1M input tokens and $1.50 per 1M output.

Key features

Coding agent

Vibe's coding agent scaffolds multi file features with tests and docs, generates unit and integration tests, opens PRs asynchronously, traces architecture across services, migrates legacy code and upgrades dependencies. Pro lists all day coding sessions in the CLI, IDE or web, and Free includes limited sessions.

Deep research and document creation

Vibe runs research across the web and internal documents and produces PDFs, presentations, reports and briefs. Libraries hold your files: 15 GB on Pro, 30 GB per user on Team.

Connectors and MCP

More than 100 connectors cover email, calendar, Slack, GitHub and Jira, available even on Free. Full MCP compatibility means you can add your own tools.

Voice input

Voice input is powered by Voxtral, Mistral's speech model.

Image generation

Free includes basic image generation; Pro raises it up to 40 times.

Enterprise deployment

Enterprise can deploy on premises or in a private cloud with full data residency control, fine tune models, white label the assistant and add audit logs and SAML SSO. No other major consumer assistant in our coverage lists self hosting.

What users say

Source	Rating	Reviews
App Store	4.8	1,300
G2	4.2	61
Trustpilot	2.4	110

The App Store rating is high but comes from about 1,300 ratings, thousands of times fewer than ChatGPT or Gemini. G2 reviewers, rating Mistral AI overall, give 4.2 from 61 reviews. Trustpilot is more balanced than for most rivals: 52% one star, 21% five star.

Recurring praise: a European alternative, coding ability and fewer restrictions. Recurring complaints: accuracy lapses on calculations and data, billing failures that block accounts without self service recovery, automated support replies and occasional slow or stuck processing.

Privacy and security

Mistral's help center says consumer Vibe data can be used to train its models by default, and that users can opt out at any time. Enterprise customers are opted out by default, with admin controls. Thumbs up or down ratings authorize Mistral to use that feedback and the related data on any plan.

Enterprise is where Mistral stands apart: on premises and private cloud deployment with full data residency control. For European companies with data sovereignty requirements, or any company that cannot send data to a US or Chinese provider, that is a deciding feature. The help article we read does not state where consumer data is stored, so check Mistral's privacy policy and data processing addendum if location matters for your plan.

Who it is for

• Budget buyers who want a real paid tier: $14.99, or $5.99 for students.
• Developers: a coding agent and $15 in API credits in one plan.
• European companies: an EU lab with on premises and private cloud options.
• Small teams on a budget: $24.99 per user with domain verification and data export.

Who should skip it

• People who want the most proven assistant: ChatGPT has millions of ratings. See ChatGPT review.
• Heavy image and video creators: see Gemini review and Grok review.
• Buyers who need to know the model on each plan: Claude and ChatGPT name theirs. See Claude review.
• Teams that want annual discounts: none listed.

Alternatives

• ChatGPT: the widest tool set, Go at $8 or Plus at $20.
• Claude: long documents and Claude Code at $20, $17 annually.
• Gemini: AI Plus at $4.99, AI Pro at $19.99 with 5 TB.
• DeepSeek: free reasoning chat and a cheap API.
• Perplexity: cited research.

Verdict

Mistral Vibe is the value pick among full featured paid assistants. $14.99 buys higher chat limits, all day coding and $15 in API credits, students pay $5.99, and enterprises get the only self hosting option among major assistants.

It falls short on proof and transparency. The review base is small, the model behind each plan is not named and Trustpilot complaints cite billing lockouts. Start on Free, and pay for Pro if the coding agent or the European data story matters to you.