Questions and answers.
The short, honest version.
What is Conifer?
Conifer is a desktop app and the engine under it: one interface for chat, agents, and coding tools, in front of every model you can reach. A router sends each query to the cheapest model that can do the job, starting with the free ones on your own computer. One account, one bill, and cloud models only when you allow them.
Is everything really local?
By default, yes: everyday queries run on your machine and your chats and files stay there. Cloud is used only when needed, and only if you allow it; privacy mode pins every query to local. Any outside call shows up on a transparency panel inside the app, so you can just check.
Which platforms can I run it on?
Mac, Windows, and Linux. On Mac it is one universal build, signed and notarized by Apple, so you will not get any scary warnings. Windows is 10 or newer. On Linux it is an AppImage, with a .deb too if you are on something apt-based. The button up top picks the right one for you automatically.
What does it cost?
Nothing to download and run: local models are the free tier. Cloud models are pay per use on one bill, only when the router needs them.
Which models does it run?
Every model. Local GGUF models are the free tier: Qwen, Llama, Gemma, downloaded once and run on your machine. Cloud and frontier models sit behind the same picker, used only when a query needs them and only if you allow them. You can set a different model for each agent.
How do updates work?
When there is a new version, the app lets you know and you decide when to install it. Whatever you have already keeps working either way, so you are never forced to update before you are ready.
Found a bug, or want something?
Tell us. Head to the feedback page and it goes straight to the people building this. We read everything.
Still curious?
The docs go deeper.