Local AI chat · Windows and Linux Ask away, at home.
ChatonLM’s chat lets you talk with a language model that runs on your own computer. You type, it answers, and nothing you say goes off to an online service.
Three steps, and you’re chatting.
Pick a model
ModelHub shows you the ones that fit your machine and installs them in one click.
Type
A question, a text to summarize, a snippet of code. Add an image or a PDF if the subject calls for it.
Read the answer
It’s written out before your eyes, formatted as it goes: headings, tables, highlighted code, formulas.
What it can do.
Images, PDFs, code
Drop a file into the window: images for models that can see, PDFs, text and code files.
A Think button
Reasoning models get their own Think button, with a thinking level when the model allows it.
A history of your own
Your conversations are kept on your computer, with favorites. Every message can still be edited.
Easy-to-read answers
Markdown, highlighted code and math formulas, rendered while the answer is being written.
The web, if you want it
Connect an Ollama account, and the model can search the web. It’s optional: without an account, nothing goes out.
A model that’s always ready
Keep a model in memory for as long as you like, or load it when the app opens so the first answer doesn’t have to wait.
Your conversations stay home.
Online AI services receive everything you write to them. Here, the model runs on your computer: your messages, attachments and history never leave it. Two exceptions, which are yours to decide: web search, handled by your Ollama account, and an Ollama server you may have set up on another machine.
The details, connection by connectionQuestions about chat
Does it work without the internet?
Yes, once the model is downloaded. The chat happens entirely on your computer.
Do I need a graphics card?
No. Models also run on the processor, just more slowly. With a graphics card they answer faster and can be bigger.
Which models can I use?
Models you can download and run locally, such as Gemma, Qwen, Mistral or Llama. ModelHub tells you which ones fit your machine.
Is it as capable as the big online AIs?
Not at everything. A model that fits on your computer is smaller than the ones behind the big online services. For chatting, summarizing, proofreading, translating or studying, it does the job very well, and your data stays with you.
Do I need to create an account?
No. ChatonLM has no account and no subscription. An Ollama account is only needed if you want web search.