Opisense BotSelf-hosted
Your servers. Your models. Your audit log.
Run Opisense Bot on infrastructure you control. Postgres, bot files, browser profiles and the audit log stay inside your deployment.
What stays with you
Your data
Postgres, bot files, browser profiles and the audit log stay on the deployment you run.
Your models
Use a hosted provider with your own keys, an OpenAI-compatible endpoint or a local model. Each bot can use a different one.
Your compute
Bots work in Docker computers on your server by default. A single machine needs no third-party sandbox account.
Bring the model and compute you trust
Models
- OpenAI
- Anthropic
- OpenRouter
- OpenAI-compatible
- Ollama
- LM Studio
- llama.cpp
- MLX
Bot computers
- Docker
- E2B
- Daytona
- CreateOS
- Boat
Runs where you need it
A single server
One installer pulls the published images, generates secrets and starts the stack with Docker Compose.
A production VM
Put HTTPS in front, point the desktop and mobile apps at your address, and bots keep working around the clock.
Restricted networks
Mirror the installer downloads and pair them with a local model so prompts stay on your network.
Self-hosted, not on your own
The team that builds Opisense Bot helps you plan, harden and maintain your deployment.
- Architecture review for your environment
- Hardening, secrets and backup guidance
- An upgrade path as new releases ship
- A direct line to the product team
Questions
Plan your deployment.
Contact us to set up a self-hosted deployment. We’ll walk through the architecture with your team.