Is there a European ChatGPT? What it means, and the options
Yes: a European ChatGPT is an AI chat run by a European company on servers in Europe, rather than by a US group. Mistral Vibe (formerly Le Chat) in France and Lumo by Proton in Switzerland are two of them. Evanesca, a private AI agent and chat published by a French company, is another: it runs 16 models from several makers in Europe and keeps no copy of your messages. They differ on what they keep, whether they train on your chats, and which models they offer.
Facts checked on 6 Nov 2026
What people mean by a “European ChatGPT”
People who look for a European ChatGPT usually want the same thing — a capable AI chat — with one change: their words should stay in Europe, under European law. That breaks down into four questions, and a service can pass one and fail another.
- The company. Who you contract with, and who owns it. The European subsidiary of a US group is still part of a US group.
- The place. Where the model runs when you send a message: a country or a region the service names, not a vague “hosted in Europe”.
- The law. The GDPR applies to any service that processes the data of people in the EU, ChatGPT included. It is the floor, not what makes a service European.
- The reach of US law. The CLOUD Act lets US authorities require a provider under US jurisdiction to disclose data in its “possession, custody, or control”, wherever it is stored source. A data centre in Paris or Frankfurt does not change that when the provider is a US company.
Two more questions matter as much in daily use: how long your conversations are kept, and whether they train the next model. GDPR vs CLOUD Act explains why EU hosting alone is not enough, and the sovereignty page shows how Evanesca's operators were checked.
Is there a European version of ChatGPT?
Not for individuals. In the EU you contract with OpenAI Ireland Ltd, in Dublin source, but on the plans for individuals, content is stored “on OpenAI systems and our trusted service providers’ systems in the US and around the world” source. OpenAI offers data residency in Europe to “eligible API customers and new ChatGPT Enterprise/Edu customers” source — and wherever the data sits, OpenAI remains a US group.
Conversations are kept until you delete them, then for up to 30 days source, and OpenAI may use them to train its models unless you opt out source. The line-by-line comparison is on Evanesca vs ChatGPT; is OpenAI GDPR-compliant? covers the legal side.
European ChatGPT alternatives, side by side
ChatGPT is in the first column for reference. Each cell links to its source: OpenAI's, Mistral's and Proton's own pages, as read on 6 October 2026 for our comparisons, and Evanesca's pages for Evanesca.
| Question | ChatGPT | Mistral Vibe | Lumo | Evanesca |
|---|---|---|---|---|
| Company | OpenAI, a US group; OpenAI Ireland Ltd in the EU source | Mistral AI, Paris, France source | Proton AG, Geneva, Switzerland source | FLEECE AI SASU, France legal notice |
| Where your words are processed | “In the US and around the world” source | In the EU by default; some features may move data outside the EU for a time source | “Proton’s fully European infrastructure”; no country named source | In Europe only: France, Mistral AI's EU region, Germany and Switzerland sovereignty |
| Conversations kept | Until you delete them, then up to 30 days source | Until you delete them, for your history source | Prompt erased after the answer; history encrypted so only you can read it source | No copy kept by Evanesca; history in your browser privacy |
| Training on your chats | Possible unless you opt out source | On by default, unless you opt out source | No; anonymised feedback only if you choose source | Never by Evanesca privacy |
| Models | OpenAI's own source | Mistral's own source | Qwen 3.5, GLM 5.3 and Apertus 1.5, picked for you source | 16 models from several makers, picked by you or by Auto mode models |
| Where you use it | Web, iOS and Android source | Web, iOS and Android source | Web, iOS and Android source | In the browser, and through an OpenAI-compatible API docs |
| How you pay | Free plan; Go, Plus and Pro subscriptions source | Free plan; Pro and Team subscriptions source | Free, even without an account; Lumo Plus subscription source | Per answer, from €5 of credit, no subscription required pricing |
Other private chats often come up in the same searches — Venice AI, Kagi Assistant, Brave Leo — but they are run by US companies: Venice.ai's terms are under Wyoming law source, and Kagi and Brave are listed with their companies on the sources page. Evanesca vs Venice AI compares the first.
Mistral Vibe, formerly Le Chat: the French ChatGPT
Mistral AI is a French company, registered in Paris source, and its assistant is the closest European match for ChatGPT's scope. Mistral's help centre puts the renaming plainly: “Le Chat is now Vibe”, with modes for multi-step work, for code and for turn-by-turn chat source. It runs Mistral's own models, with apps on the web, iOS and Android source, and “by default, your data is hosted in the European Union”, although some features can move it outside the EU for a time source.
The trade-off is retention and training. Vibe keeps your conversations to show your history — zero data retention “is not available for Vibe Work or Chat” source — and uses inputs and outputs to train Mistral's models by default, unless you opt out; Enterprise customers are opted out by default source. Evanesca vs Mistral Vibe compares the two in detail.
Lumo by Proton: the Swiss option
Lumo is run by Proton AG in Geneva, a Swiss company overseen by the non-profit Proton Foundation source, on what Proton calls its “fully European infrastructure, protected by Swiss privacy laws” source. Switzerland is in Europe but not in the EU; the European Commission recognises its data protection as adequate, so data can flow there from the EU without extra safeguards source.
« After Lumo processes your query and generates a response, the data is erased. »
Lumo keeps no logs and no metadata, stores your history encrypted so that only you can read it, and does not train on your chats; it picks the model for you among a few open ones source. Evanesca vs Lumo goes through both.
Evanesca: a European ChatGPT alternative that keeps no copy
Evanesca is a private AI agent and chat, published by FLEECE AI, a French company. It runs 16 models from Mistral AI, the Swiss AI Initiative, Qwen, DeepSeek, Z.ai, OpenAI, Meta, Google, Moonshot AI and Aleph Alpha, and every one of them runs in Europe: in France by OVHcloud and Scaleway, in Mistral AI's EU region, in Germany by AKI.IO and Verda, and in Switzerland by Infomaniak. Evanesca serves no model outside Europe. OVHcloud, Scaleway, Mistral AI and Infomaniak are European companies with no US parent, out of the CLOUD Act's direct reach; AKI.IO and Verda are European companies whose shareholders are not all published.
Your prompt and the answer are held in memory while the answer is written, then dropped; your history stays in your browser, and of your messages only billing data — tokens and cost — is kept. Evanesca never trains a model on what you type. The operators state their own part: Scaleway may keep and read requests that cause errors or look abusive, for up to 14 days source; Mistral AI keeps no input or output beyond the time needed to answer, as Evanesca's account has been on zero data retention since 17 November 2026 source, and lets its customers object to training — Evanesca's account objects source source. The passport under each answer names the model, the operator, the country and what was kept.
Beyond chat, Agent mode carries out a task step by step — it reads the conversation's files, writes Word, PDF and Excel documents and creates images — within a spending cap you set, and never runs code. Web search is off by default; when you switch it on, each search is billed, and Linkup, a French company, receives only the query the model writes and the addresses of the pages it reads — never the whole conversation source. You pay per answer, VAT included, from €5 of credit, with no subscription required pricing. What makes a chat private, beyond where it runs, is on private AI.
Where ChatGPT is still ahead, and how to choose
- ChatGPT has OpenAI's frontier models and more features — voice, scheduled tasks, custom GPTs, Codex source. OpenAI does not publish those models, so Evanesca does not have them; the one OpenAI model it runs is the open-weight GPT-OSS 120B.
- Mistral Vibe has Mistral's whole range of models, connectors to your tools and a coding agent source; Evanesca serves a few of Mistral's models, next to other makers'.
- Lumo goes further on encryption and syncs your history across devices source; Evanesca keeps your history on the device you use.
- ChatGPT, Vibe and Lumo all have mobile apps and a free plan; Evanesca runs in the browser and has no free plan — iOS and Android apps have been submitted to the App Store and Google Play; they are not published yet (6 November 2026).
So choose by what matters most. Mistral Vibe if you want one French assistant for everything and are comfortable with your history being stored. Lumo for encrypted private chat, synced across devices, under Swiss law. Evanesca if you want no stored copy, a choice of models run in Europe, an Agent mode and a record of who processed each answer.
Frequently asked questions
Is there a European ChatGPT?
Yes, several. Mistral Vibe, formerly Le Chat, from the French company Mistral AI, is the closest in features. Lumo, from Proton in Switzerland, focuses on encrypted private chat. Evanesca, from a French company, runs models from several makers in Europe and keeps no copy of your conversations. ChatGPT itself has no European version for individuals.
Is Mistral the French ChatGPT?
Mistral AI is a French company, and its assistant — Le Chat, now called Vibe — is the nearest French equivalent. By default it hosts your data in the EU; it keeps your conversations for your history and uses them for training unless you opt out.
Does ChatGPT store my data in Europe?
Not on the plans for individuals: OpenAI stores content in the US and around the world. It offers data residency in Europe only to eligible API customers and new ChatGPT Enterprise and Edu customers.
Is an AI chat hosted in the EU safe from the CLOUD Act?
Not if the provider is a US company: the CLOUD Act lets US authorities require a provider under US jurisdiction to disclose data it controls, wherever it is stored. What counts is the company as well as the place.
Is Evanesca free?
No, there is no free plan. You add credit from €5, VAT included, and pay per answer at each model's public price; a subscription is optional, and credit never expires.
Read next
- The best ChatGPT alternatives, by what you needThe best ChatGPT alternative depends on your need: privacy, European law, free use, research, code or offline. Claude, Lumo, Duck.ai, Vibe and more.
- Private AI agent: what makes an agent private, and how Evanesca's worksA private AI agent carries out tasks without keeping your data. What to check — where it runs, what it keeps, what it can touch — and Evanesca's Agent mode.
- Private AI: what it means, and how to judge onePrivate AI keeps your words from being stored, read or used for training. The three kinds, a checklist to judge any of them, and the main options compared.
- European AI: who makes it, who runs it, and how to use itWhat European AI covers: models made in Europe, European operators and data centres, the AI Act — which ones Evanesca runs, and how to use them today.
- Sovereign AI: what the word covers, and how to check a claimSovereign AI means control of the model, the servers, the operator and the law. A checklist to judge any claim, and where Evanesca stands on each layer.
- AI for lawyers and law firms: what it is good for, and how to keep client secretsAI for lawyers: drafting, summaries, translation, contract review — and client confidentiality. What the French bar says, a checklist for law firms.