Run local, private AI in the free Eclipse IDE

Not every team can send code to a cloud API. Eclipse lets you run models locally with Ollama and OpenAI-compatible endpoints, so your prompts and source stay on your own machine — online or fully offline.

Why run AI locally

Data stays put

Code and prompts are processed on your device. Nothing is sent to a third-party provider when you use a local model.

Works offline

A local model needs no internet and no account — useful on air-gapped or restricted networks.

No usage bill

Local inference has no per-token cost, so high-volume tasks stay affordable and predictable.

Mix local and cloud

Keep a local model for sensitive code and quick edits, and a cloud model for the hardest problems. Eclipse lets both live in the same setup, so you choose per task without leaving the IDE.

See all providers →
# local model via Ollama
provider: ollama
endpoint: "http://localhost:11434"
model: "your-local-model"
 
# no key, no cloud, no data leaving

About local AI

Can I use Eclipse AI completely offline?

Yes. With a local model served through Ollama or an OpenAI-compatible endpoint, the AI features run on your own machine and do not need an internet connection or a cloud account.

Does my code leave my machine with local models?

No. When you use a local model, prompts and code are processed on your device, so nothing is sent to a third-party provider. This suits private, sensitive or regulated work.

Can I mix a local model with a cloud model?

Yes. You can keep a local model for sensitive code and a cloud model for harder problems, and switch between them per task without leaving the IDE.

Private AI, your machine

Download Eclipse free and run a local model in minutes.

Download free for Windows