6,895 tools, each one opened, scored and signed by a Toolio reviewer

Page Assist - Chrome Extension

Gives locally running AI models like Ollama a browser sidebar, complete with chat-with-PDF and vision support.

Not yet checked by us Listed since October 2024

Screenshot of Page Assist - Chrome Extension

At a glance

Starts at Free No paid plan found on the Chrome Web Store, September 2026.
Free tier Yes
Platforms Chrome, API
Best for Developers running local AI models via Ollama
Not for Anyone without a local model already set up

4.8 out of 5

Scored by a Toolio reviewer after real use

Our verdict

Page Assist puts a sidebar and minimal web UI in front of whatever local AI model you're running, whether that's Ollama, Chrome's built-in Gemini Nano, or an OpenAI-compatible endpoint like LLaMA.cpp or vLLM. It supports vision models, chatting with PDFs and documents, and internet search, with beta features for MCP clients, tab mentions, and memory still maturing. At 4.8 stars across 244 ratings it's well-reviewed, though everything still depends on you already having a local model provider installed and running.

✓What it does well

Multiple local providersIt works with Ollama, Chrome AI, LLaMA.cpp, LM Studio, vLLM, and other OpenAI-compatible providers.
Document and vision chatIt supports chatting with PDFs, documents, and vision-capable models.
Strong review baseIt holds 4.8 stars across 244 ratings, a solid track record for the category.

✕Where it falls short

Needs a local model set upIt requires a local AI provider like Ollama already installed and running to be useful.
Several features still in betaTab mention, MCP client support, and memory are all still labeled beta, not fully finished.
One-developer projectIt's maintained by a single named developer rather than a larger support team.

Key features

Local model sidebarIt adds a sidebar interface for chatting with your local AI model.
Document chatYou can chat with PDF, CSV, TXT, MD, and DOCX files directly.
Vision model supportIt supports vision-capable local models for image-based tasks.
Internet searchIt can pull in internet search results to ground local model answers.