ControleveryAIrequestbeforeitleavesthebuilding.
One gateway for every AI model in your organization. Allocate cost by team, route each request to the right model, and stop sensitive data from leaving.
Requests come in
One gateway decides
The best model answers
Allocate & control cost
See exactly what every team, person, and app spends on AI — and set budgets before they overspend.
Use the right model
Automatically route each request to the most cost-effective model that meets the task and quality bar.
Keep data in-house
Enforce policy on what information can leave your organization, by team, app, or data type.
Tandevia Gateway
Every AI request in your organization — from employees, apps, or agents — routes through one governed layer. It's where cost, model choice, and data policy are enforced, no matter which vendor sits behind it.
Unified API
One endpoint for every model provider — OpenAI, Anthropic, and self-hosted models — with instant failover.
Cost allocation & chargeback
Every request is tagged to a team, project, or app so finance can see and bill true AI cost.
Smart model routing
Policies send simple tasks to cheaper models and reserve premium models for what actually needs them.
Data loss prevention
Inspect and redact requests before they leave your network; block sensitive data by policy, not by trust.
Full audit log
Every prompt, response, model, and cost is logged and queryable for compliance and security review.
Bring your own models
Connect self-hosted or open-weight models alongside commercial providers through the same gateway.
Tandevia Assistant
The free chat app your team uses instead of ChatGPT or Claude directly — on desktop or in the browser. Vendor-independent by design: every conversation runs through your Tandevia Gateway, so switching or mixing model providers never means switching apps.
- Free, unlimited seats — every employee gets a chat app, at no cost, forever.
- Desktop and web — native apps and a browser client, same experience either way.
- Vendor-independent — switch or mix model providers behind the Gateway without anyone changing apps.
- Policy-aware by default — every chat automatically inherits your organization's routing and data rules.
Simple pricing. You never pay us for model tokens.
Tandevia Assistant is free on every tier. Gateway pricing is per human seat, plus metered throughput for the apps and agents connected to it.
- 5 users, 1 connected app
- 5M gateway tokens / month
- Tandevia Assistant included
- Usage dashboard
- 14-day log retention
- Up to 5 connected apps
- 3M tokens / user / month, pooled
- Then €0.20 per 1M app tokens
- Tandevia Assistant included
- Team budgets & alerts
- 90-day log retention
- Unlimited connected apps
- 5M tokens / user / month, pooled
- €0.12 per 1M app tokens
- Tandevia Assistant included
- SSO & per-department policies
- DLP guardrails & audit log
- 1-year log retention
- App throughput from €0.05 per 1M tokens
- Overage cap available
- On-prem deployment — runs entirely in your infrastructure
- Tandevia Assistant included
- Priority support
- Custom SLA & procurement
- Connect MCP servers with access control at app and user level
Common questions
Yes — on every tier, including Free, for as many users as you have. It's the access point we want your whole organization using instead of consumer ChatGPT or Claude accounts.
Start free
Tell us your work email. We'll send you access and reach out personally to help you get set up — no credit card needed.
By continuing you agree to be contacted by our team by email.
Book a demo
Tell us a bit about your team and we'll get in touch to schedule a walkthrough.
This just collects your details — our team follows up by email.
Bring every AI request under one roof.
Start with free Assistant seats today. Add the Gateway when you're ready to govern spend and data.