A browser with a built-in AI assistant

What “built-in AI” really means here

A browser with built-in AI is not a chatbot you paste links into. LLM Browser is a full desktop web browser, with tabs, an address bar, bookmarks, history, downloads, light and dark themes, and a settings panel, just like the browser you use today. The difference is the assistant living in a docked bridge panel in the same window. It can see the live page and act on it directly.

Because the assistant is part of the browser rather than an add-on, it works on the real pages you are already signed in to. There is no copy and paste, no separate automation tool, and no hidden tab you cannot follow. You watch each step happen in the panel as it runs.

What the assistant can do on the page

Once you connect an AI, the assistant becomes a hands-on co-pilot that works the page in front of it:

• Navigate to pages and follow links for you.
• Click buttons and links on the live page.
• Fill and submit forms, from a single field to a long sign-up flow.
• Scroll long pages and switch tabs as a task needs.
• Read and summarise what is on screen, so answers stay grounded in the actual page.
• Take screenshots of what it sees while it works.

In practice that means asking it, in plain words, to find software developer jobs in a city, triage an inbox, or gather details across several pages, then watching it do the clicking and typing.

You choose the AI, including private and offline

A built-in assistant is only useful if you can pick the model behind it. LLM Browser gives you six connection types in one LLM Connection settings tab: an API Key for providers such as OpenAI, Anthropic, DeepSeek, and Groq, an OAuth sign-in, a Local Server such as Ollama or LM Studio, a built-in Local AI GGUF model file for fully offline use, a CLI Pipe, and a Terminal command. Run a local model and your browsing stays on your own machine.

Each connection carries its own reasoning level, system prompt, and step limit, so you can keep a small local model quick and focused or let a larger model take on longer multi-step work. Read more about running it offline on our guide to the AI browser with a local LLM.

Setting it up takes one extra step

Because LLM Browser behaves like a normal browser out of the box, the only setup beyond installing it is telling it which AI to use. Install it like any desktop browser, open a few tabs, sign in to the sites you use, and set your homepage and theme in Settings → Browser. Then open Settings → LLM Connection, pick one of the six connection tabs, fill it in, and press Test. A green Connected banner confirms the assistant is ready. From that point on you simply type into the bridge panel at the bottom of the window and watch it act.

Real chores it handles well

The point of a built-in assistant is not novelty, it is getting ordinary web work done with less clicking. Common examples include:

• Job hunting: search a board, surface roles that match, and open the promising ones in tabs.
• Inbox triage: read through messages, clear obvious spam, and flag what needs a reply.
• Research gathering: pull the key points from several pages and report them back in one place.
• Repetitive entry: fill the same kind of form across a run of pages without you retyping every field.

In each case the assistant works the live page, you watch it happen, and you can step in whenever you want to.

Why a built-in assistant beats an extension

Extensions are bolted on and usually capped at a sidebar chat that cannot truly act for you. Because the assistant here is part of the browser, it reads and operates the same page you are looking at, with you in the loop the whole time. If you want to go further, LLM Browser also exposes a local control API so your own scripts can drive the browser with an LLM, and the assistant is genuinely good at the everyday chore of an AI browser that fills forms.