EU servers
Your data stays in the European Union. Full stop. Our primary region is Supabase Frankfurt; we also operate in Amsterdam.
EchoChat in detail: how we build it
Four pillars, twenty-five features. From European privacy guarantees to the largest model list on the market:discover the engineering behind your new AI workplace.
Privacy & compliance
Built for European teams that can't compromise on data protection.
EU servers, zero training on your data, and admin controls that IT will actually approve.
Zero training on your data
Two promises: we don't store conversations for training, and every provider we route to contractually excludes your data from their training pipelines.
Local model (add-on)
Self-hosted AI that never touches the internet. For teams that need an airgap, not just a promise.
Model allowlists
Admins pick exactly which models the team can use: EU providers only, specific vendors, or a curated shortlist per role.
Upload guardrails
Detect and block sensitive data before it leaves the browser. Admins configure what counts: PII, secrets, contract language, whatever your policy says.
Incognito mode
Nothing is stored. Files and transcripts auto-wipe 30 minutes after the conversation closes.
SSO + audit logs
SAML single sign-on, role-based access, and full audit logs: the things IT asks about before signing.
Want to know more about our privacy?
Visit our privacy pageEvery model. One contract.
Every major model from every major provider. One invoice. Never locked in.
Switch mid-conversation between models. If a provider goes away, your team keeps working.
Anthropic
OpenAI
Google
Meta
Mistral
xai
deepseek
alibaba
cohere
Perplexity
moonshotai
bfl
200+ models
Claude, GPT, Gemini, Llama, Mistral, Grok, DeepSeek, Qwen, Cohere and more. Choose over 27 providers.
Flagship models on day one
When a provider ships a new frontier model, it's available in EchoChat the same day.
Image generation
Nano Banana, GPT Image, Recraft: create high-end visuals without a separate Midjourney subscription.
Video generation
Veo, Kling: text-to-video and image-to-video with controls over aspect ratio and duration.
Switch mid-conversation
Change models without losing context. Ask Claude for the analysis, switch to GPT for the rewrite.
Zero lock-in
One invoice, one contract, one place to cancel. If a provider goes away, your team keeps working.
Built for teams
AI that makes your whole team smarter: not just the one person who happened to write the prompt.
Shared chats, projects, and analytics. Visible to the people who need it, invisible to everyone else.
Share chats, images and videos
Publish a chat to your team or organization in one click. Or keep it private: your call.
Mention teammates
Tag a colleague in a conversation so they see it instantly. Great for handoffs and second opinions.
Teams and roles
Organize by department, function, or project. Give different teams different model access and credit budgets.
Project workspaces
Group chats, files, and instructions under one project. Add or remove chats as the work evolves.
Privacy-respecting analytics
Admins see usage per team and user. They never see what anyone said. Full stop.
Credit pooling (Enterprise)
Share one credit pool across the whole organization. Heavy users and light users balance each other out: nobody runs out.
Enterprise tip
Credit pooling is available on the Enterprise plan. Ideal for organizations where some teams use ten times more AI than others.
Assistants & workflows
Teach EchoChat your business once, and the whole team benefits forever.
Build custom assistants with your own knowledge. Generate images, video, code and audio: without switching tools.
Prebuilt assistant library
Ready-to-use AI assistants for common tasks: brand writer, code reviewer, meeting summarizer, data analyst.
Custom assistants
Build your own with custom instructions. Share them with your team or keep them for personal use.
Knowledge files
Upload PDFs, DOCX, CSV, JSON, or TXT as persistent context. Perfect for playbooks, product docs, and policies.
Code canvas
Live preview of generated code, inline edits, and the ability to run your changes: without leaving the chat.
Voice and audio
Speak to the AI, get audio back. Transcription and generation built in.
Mobile app
Native Android app for working on the go. iOS is on the roadmap.
Example: turn this into a LinkedIn post with a matching image
One prompt in EchoChat, and you get both the copy and the image back: no Canva, no Midjourney, no copy-paste.
Why teams switch
EchoChat vs. a pile of individual AI subscriptions
What you get when you stop juggling vendors and start working in one workspace.
| Capability | Six separate subscriptions | EchoChat |
|---|---|---|
| Cost visibility | Five invoices, three currencies, manual reconciliation | One invoice, usage analytics per team |
| Team sharing | Copy-paste chats into Slack | Share with one click, @mention teammates |
| Admin controls | Hope people don't share sensitive data | Model allowlists, upload guardrails, audit logs |
| EU hosting | Mostly US-hosted, scattered data residency | EU-hosted by default, contractual no-training |
| Custom assistants | One per vendor, can't share between teams | Unlimited, org-wide sharing, with knowledge files |
| Vendor lock-in | Stuck. Switching means migrating everything. | Switch providers mid-conversation. No lock-in. |
Frequently asked questions
What teams usually want to know
Replace your pile of AI subscriptions with one secure workspace.
Join the teams using EchoChat to bring Claude, GPT, Gemini and 200+ other models under one invoice.