Page Assist
Page Assist is a tool designed to augment web browsing through the use of locally running AI models. It serves as a Web User Interface (UI)...
Last verified:
What is Page Assist?
Page Assist is a browser extension that provides a sidebar and web‑style UI for your locally running AI models, enabling you to interact with them directly from any webpage. It lets you use models served by providers such as Ollama, Chrome AI (Gemini Nano), and any OpenAI‑compatible API endpoint like LLaMA‑cpp, LM Studio, or vLLM, all while keeping your data and queries local or on‑prem as much as possible. The extension is designed to act as a lightweight assistant during browsing, supporting tasks such as summarizing page content, answering questions about the page, and chatting with documents you upload.
Key features include an always‑visible sidebar for quick access, support for vision models, and a minimal but functional chat interface similar to a web‑based LLM chat. You can chat with PDFs, CSVs, plain text, Markdown, and DOCX files directly in the sidebar, and the tool also supports internet search, tab mentions, an experimental MCP client, and memory functions to carry context across sessions. The UI is intentionally lean, focusing on fast, low‑overhead interaction with your own models rather than heavy‑duty workspace features.
Page Assist is mainly aimed at developers, privacy‑minded users, and AI enthusiasts who want to integrate local or self‑hosted LLMs into their everyday browsing without relying on cloud‑only assistants. It fits well for people already running Ollama or similar tools and who want a simple, always‑on sidebar chat instead of switching between a terminal, desktop app, or full‑blown web UI. Because it is open‑source and free, it is also attractive for experiments, custom workflows, and integration into local‑only AI setups where external web‑based services are avoided.
The tool emphasizes simplicity and local‑first behavior: you install the extension, point it to your running provider (e.g., Ollama or an OpenAI‑compatible server), then use the sidebar to chat with the model and analyze both the page and uploaded files. It does not aim to be a full‑fledged AI IDE or project environment but rather a focused companion for browsing‑adjacent tasks such as quick research, code explanations, and document‑based Q&A.
Page Assist pricing
Pricing model: Free
Page Assist is a completely free, open‑source browser extension with no paid plans, subscriptions, or premium tiers. It is offered as gratis software on the Chrome Web Store and other browsers, and the developer states that it will remain free to use without hidden costs for the core functionality. Users still pay indirectly for computational resources by running their own local AI models or using their own OpenAI‑compatible API endpoints, but the extension itself does not charge for access, features, or usage limits.
Page Assist pros
- Runs as a browser sidebar so you can use it on any page
- Works with locally hosted Ollama models for privacy
- Supports Chrome AI (Gemini Nano) for local‑on‑device use
- Connects to any OpenAI‑compatible API provider like LLaMA.cpp or LM Studio
- Minimal web UI keeps overhead and latency low
- Supports vision models for image‑based queries
- Chat with PDFs directly in the sidebar
- Chat with multiple document types including CSV, TXT, MD, and DOCX
- Includes built‑in internet search for up‑to‑date results
- Beta Tab Mention feature to reference current page content
- Experimental MCP client support for additional tooling
- Memory support that preserves context across chats
- Always‑free, no pay‑per‑use or subscription model
- Open‑source so you can inspect and self‑host the extension
- Lightweight and small download size for an AI extension
- Available on multiple browsers including Chrome, Firefox, and Edge
Page Assist cons
- Requires you to set up and maintain your own local AI stack (e.g., Ollama)
- No built‑in model hosting; you must run your own provider
- Some features are explicitly marked as beta and may be unstable
- Limited advanced project management or multi‑workspace features
- No native mobile app; only browser‑extension workflows
- Performance depends heavily on your local hardware and model size
- No official mobile‑browser support details provided
- Smaller ecosystem compared to big‑cloud AI assistants
Frequently asked questions about Page Assist
What does Page Assist do?
Page Assist is a browser extension that adds a sidebar and web‑style chat UI for your locally running AI models, letting you interact with them while browsing. It can connect to local providers like Ollama or Chrome AI, as well as any OpenAI‑compatible server, and you can use it to ask questions about the page you are on, summarize content, or chat with documents you upload directly in the sidebar.
Which AI providers does Page Assist support?
Page Assist supports Ollama, Chrome AI (Gemini Nano in beta), and OpenAI‑compatible API endpoints such as LLaMA‑cpp, LM Studio, Llamafile, vLLM, and other self‑hosted or open‑source servers that expose an OpenAI‑style API. You configure the extension to point to your chosen provider’s local or on‑prem endpoint and then use the sidebar chat to interact with the models exposed there.
Do I need to pay to use Page Assist?
No, Page Assist is completely free to use with no paid plans, subscriptions, or feature‑locked tiers. The extension itself costs nothing to download or install, although you may incur compute or hosting costs from running your own AI models or using an external API provider that charges per token or per request.
Can it work with PDFs and other documents?
Yes, Page Assist lets you chat with PDFs and several other document types directly in the sidebar, including CSV, TXT, Markdown (MD), and DOCX files. You can upload these files, and the extension will send them to your configured local AI provider so you can ask questions, extract data, or summarize content without leaving the browser.
Is Page Assist open source?
Yes, Page Assist is an open‑source browser extension whose code is hosted on GitHub. This allows you to inspect how it works, contribute to its development, or build and load a custom version of the extension from source if you prefer not to use the store‑published build.
How does Page Assist handle privacy and data?
The developer states that Page Assist does not collect or use your data for purposes unrelated to the extension’s core functionality and that your data is not sold to third parties. Most of the AI processing happens locally or on your own provider, so your queries, documents, and page content stay under your control, but you should still review the specific privacy policy of any external AI endpoint you connect to.
Does Page Assist need an internet connection to work?
Page Assist can work offline if your chosen AI provider (such as a local Ollama instance) is running on the same machine and does not require external internet access. However, if you enable internet search or connect to a cloud‑based OpenAI‑compatible API, then an internet connection is required for those features to function.
What features are still in beta?
The extension lists Tab Mention, MCP client support, and memory as beta features. Tab Mention lets you refer to the content of the current tab in your chat, MCP client support adds extra tool‑integration capabilities, and memory aims to preserve conversation context across sessions; because these are labeled beta, they may change or behave unreliably in some setups.
Which browsers can use Page Assist?
Page Assist is designed for Chromium‑based browsers such as Chrome, Brave, and Edge, and it also supports Firefox. You can install it from the Chrome Web Store for Chrome and Edge, from the Firefox Add‑ons site for Firefox, and you can build and load it manually for other Chromium‑derived browsers.
How do I install and configure Page Assist?
You install Page Assist as a browser extension from its store listing, then open the sidebar and configure your local AI provider (such as Ollama or an OpenAI‑compatible endpoint) by entering the provider’s URL and any required API key. After that, you can start chatting with your models, upload documents, and use the sidebar on any webpage to interact with your local AI while browsing.