Bring your own model. Tell us how you want to leverage AI across your operations, and we'll show you how CELERiS puts AI to work - ensuring complete data sovereignty without tying you to a single vendor, a single cloud, or even a connection.
Unlock the power of LLMs and generative AI with a seamless layer that works across any model or provider.
One completion API over many providers. Move between hosted and local models without rewriting a line of application code.
Use a hosted frontier model, or keep everything in-house with local inference - the same abstraction either way, so sensitive data never has to leave your walls.
Token accounting and cost tracking are part of the layer, so AI spend stays visible and controllable across every backend.
Both the hosted Claude API and the Claude CLI - use the frontier model however you already access it.
Gemini via the CLI for fast, capable hosted completions.
Enterprise-grade hosted inference on Google Cloud's managed platform.
Run open models locally with a single command - great for development and air-gapped deployments.
Lightweight, on-device inference - local-first AI down to modest hardware.
The abstraction is open by design - adding another backend is implementing one trait, not re-plumbing the platform.
An AI-powered provider that discovers new transport data sources and generates the scrapers to ingest them - shrinking the work of bringing a new operator, region or mode online into the OpenJourney model.
CELERiS exposes an MCP server, so AI agents can query and act on live journey data through a current, composable, standard interface - ready for the AI-agent era rather than locked to one assistant.
Any LLM backend, hosted or local - this page.
Any data source as a plugin or out-of-process runner.
One storage API over local disk or any cloud, deployable from a Pi to Kubernetes.
Tell us how you want to leverage AI. We can help you put provider-independent models to work on your moving world.