Run Remy on your own LLM
A new way to use Remy, in development now. Point it at a model running locally on your machine, or any API LLM service you choose — for total control over where your data goes.
The same buddy, with the keys in your hands
The Mac app keeps things simple. The CLI goes the other way — every decision about your data is yours to make.
Your data, your machine
Run a fully local model and nothing ever leaves your computer. No accounts, no clouds, no middle-man — the strongest privacy Remy can offer.
Bring any model
Point Remy at a local model (like the ones you run with Ollama or LM Studio) or any API LLM service you already pay for. You pick the brain.
Built for the terminal
A scriptable command-line companion for people who live in the shell — same screen-aware, step-by-step Remy, no window required.
You decide what the model is, and where it runs
With the Mac app, Remy runs on your own Claude account. The CLI widens that choice all the way: keep it 100% local, or plug in the API provider you trust. Either way, there’s no Remy proxy in the middle.
- Choose a local model — everything stays offline on your machine
- Or connect any API LLM provider you already use
- Decide exactly what Remy can see and which tools it can run
- No Remy server in the path — your prompts go straight to your model
# run Remy on a local model
$ remy --model local
↳ nothing leaves your machine
# or bring your own API LLM
$ remy --provider your-llm-service
↳ your key, your account, no markup
Commands shown are illustrative — the CLI is still in development.
Want the CLI when it lands?
Join the waitlist and we’ll let you know the moment the bring-your-own-LLM CLI is ready to try.