A consistent local endpoint
Configure compatible clients against the gateway on the same computer instead of maintaining several scattered model URLs.
Local AI routingWindows release candidate
EasyRoute
One local gateway. Your own models.
A Windows gateway that brings installed Ollama models into a clear local routing workflow.
Ask about EasyRouteWhat it is for
EasyRoute connects compatible clients to AI models running on your own Windows computer. It discovers local Ollama models and exposes a consistent gateway, so the model endpoint is easier to configure and understand.
It is designed for people who want a deliberate local setup: install the runtime, choose a model that fits the machine, start the gateway and connect a compatible client. Processing depends on your computer and the model you install.
Read the product detail before committing. Requirements and boundaries are part of the product, not small print.
Designed around the work
Discover installed models
Use models already installed in Ollama, with named local and automatic routes for the supported workflow.
Keep routing local
The supported configuration uses local Ollama models. It does not silently send work to a paid cloud model.
Understand the setup
A small, inspectable tool with documented runtime requirements makes ownership and troubleshooting easier.
What you need
- A Windows computer with enough memory for your chosen model.
- Node.js 22 or later.
- Ollama installed with at least one downloaded local model.
- A compatible client and internet access for the initial dependency/model installation.
A good fit for
- Independent developers exploring local AI.
- Owners who want predictable local model routing.
- Windows workspaces with a compatible client and an installed model.
Typical workflow
- Install Node.js, Ollama and a suitable local model.
- Start EasyRoute and confirm the local model is available.
- Point a compatible client at the documented local gateway.
- Run a small real request before adopting a larger workflow.
Boundaries to know
- The current package is a Windows release candidate; ask about installation validation for your machine.
- Hosted Ollama models and cloud fallback are outside this local configuration.
- Model quality and speed depend on the model and available hardware.
Talk it through