Deployment
Our cloud in your region, or your own servers.
Start fast in DeepVox cloud, in a dedicated environment hosted in your region. Or run the same platform on your own servers or private cloud, operated by your team or by us.
- Options
- DeepVox cloud, private cloud, on-premise
- Cloud regions
- United States, Europe, India
- Isolation
- A dedicated environment per client
- Models
- Ours by default; outside providers only if you opt in
In short
DeepVox runs in three ways. In DeepVox cloud, each client gets a dedicated environment hosted in its own region, in the United States, Europe or India, priced per user. Self-hosted, the whole system, including connectors, index, models, voice and logs, runs in the client’s data centre or private cloud account for a one-off implementation cost, with unlimited users. In every option the models are DeepVox’s own by default, and outside AI providers are only used if the client switches them on.
Video walkthrough Coming soon
See DeepVox for Deployment in action
The walkthrough video for this page is in production. Want to see it sooner? We’ll show you live, on content like yours.
Options
Three ways to run it.
DeepVox cloud
We host and run it in a dedicated environment in your region. Nothing to install, per-user pricing, the fastest start.
DeepVox cloud →Private cloud
Your own cloud account and region, under your keys and policies. Unlimited users.
Private cloud →On-premise
Your data centre and GPUs. Nothing leaves your network. Unlimited users.
On-premise AI →Side by side
Which one fits.
| DeepVox cloud | Private cloud | On-premise | |
|---|---|---|---|
| Who runs it | DeepVox | You, or us under managed support | You, or us under managed support |
| Where data lives | A dedicated environment in your region: US, EU or India | Your cloud account and region | Your data centre |
| Models | Ours by default; outside providers only if you opt in | Same | Same |
| Pricing | Per user | One-off implementation cost, unlimited users | One-off implementation cost, unlimited users |
| Time to start | Fastest: no infrastructure to set up | Weeks, depending on your cloud set-up | Depends on hardware |
| Source code | Runs on the DeepVox platform | Yours from the first commit | Yours from the first commit |
| Best for | Teams that want to start fast without running servers | Organisations standardised on their own cloud | Strict sovereignty or professional-secrecy needs |
You can start in DeepVox cloud and move to your own infrastructure later. The platform, connectors and configuration are the same.
What gets deployed
Every layer, in your environment or ours.
| Layer | Typical components |
|---|---|
| Ingestion | Docling, Unstructured, Apache Tika, Tesseract |
| Index and vector store | PostgreSQL with pgvector, Qdrant, OpenSearch |
| Model serving | vLLM, SGLang, Ollama |
| Model gateway | LiteLLM, with usage and cost per department |
| Identity | Keycloak, Open Policy Agent, Entra ID integration |
| Observability | Langfuse, Phoenix, Grafana stack |
| Orchestration | n8n, custom services, Kubernetes and Helm |
Operated your way
Run it yourself, or let us.
- DeepVox cloud: hosted, patched and monitored by us, in your region.
- Runbooks and documentation as contracted deliverables.
- Training for your operations team.
- Managed support with a 99.9% uptime SLA, if you want it.
- A rehearsed exit test so a successor supplier can take over.
FAQ
Questions, answered.
What hardware do we need?
It depends on the models, users and whether you run voice; we size it in discovery. Small fine-tuned models keep the footprint modest.
Can we start without buying GPUs?
Yes. Start in DeepVox cloud, in a dedicated environment in your region, and move to your own infrastructure later without rebuilding.
Is our data mixed with other clients’ data in DeepVox cloud?
No. Each client gets its own dedicated environment, hosted in its region: the United States, Europe or India.
Does DeepVox cloud send our data to OpenAI or other AI providers?
Not by default. DeepVox cloud runs our own models in your region. Outside AI providers are only used if you switch them on, for the uses you choose.
Who patches and upgrades it?
Your team with our runbooks, or us under managed support.
Related
Security
Controls and data handling.
Learn more →Models of your choice
Open-weight, fine-tuned or hosted.
Learn more →Open-source stack
Every layer and the tools we use.
Learn more →Start with two weeks of evidence, not a sales call.
A fixed-price discovery sprint, credited against whatever comes next. Or write to sales@deepvox.ai.