Page Assist - A Web UI for Local AI Models

why could I not pull models? I have use cloud studio to run ollama with deepseek-r1:8b and llama3:latest. But the url added to page assist shows ollama could run, but without models! I do have deepseek-r1:8b and llama3:latest, how could that possible? Could anyone to solve the problems? Thx!

I really wanted to like this extension because the interface is genuinely promising and there are a lot of settings and configuration options. Unfortunately, it doesn’t support actually performing actions on web pages — no clicking elements, no following links, no navigation, no real interaction with the page. That makes it effectively useless if your goal is anything beyond reading or chatting about content. A web “assistant” that can’t take basic actions on the page provides zero practical value, regardless of how polished the UI looks. If future versions add true page interaction (clicking, link navigation, form actions), I’d reconsider. As it stands, it’s all configuration and no execution.

its neat but very slow same hardware running openweb ui can do the tasks 100 times faster

Works awesome, but why no option to summarize YouTube videos?! That's essential!

ONE PROBLEM - Web Search is Very Slow and there is need to Add DuckDuckGo API option for AI's search which is also Free for Unlimited Requests and is Fast too, Only This is Needed Otherwise Its best Lightweight Offline AI Extension.

Access to fetch at 'http://127.0.0.1:11434/api/tags' from origin 'chrome-extension://jfgfiigpkhlkbnfnbobbkinehhfdhndo' has been blocked by CORS policy: The 'Access-Control-Allow-Origin' header has a value 'http://127.0.0.1' that is not equal to the supplied origin. Have the server send the header with a valid value, or, if an opaque response serves your needs, set the request's mode to 'no-cors' to fetch the resource with CORS disabled.