TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Show HN: MiniSearch, a minimalist search engine with integrated browser-based AI

17 点作者 felladrin超过 1 年前
Hey everyone!<p>I’m excited to announce the release of my last project, MiniSearch.<p>I admire Perplexity.ai, Phind.com, You.com, Bing, Bard and all these search engines integrated with AI chatbots. And as a curious developer, I took the chance and created my own version.<p>Using Web-LLM and Transformers.js to provide browser-based text-generation models on desktop and mobile, I built a minimalist self-hosted search app on which an AI analyses the results, comments on them and responds to your query summarising the info. In the backend, it still queries a real search engine, but besides that, there&#x27;s no other remote connection happening.<p>For running in the browser and on mobiles, lightweight models are required, so we can&#x27;t expect them to give stellar answers, but there are a few advantages of using this over the services as mentioned earlier:<p>- Availability: The AI will always be available and respond with the maximum available speed from the device. - Privacy: Besides the queries that go anonymously to the actual search engine, nothing else leaves your device. - No ads&#x2F;trackers: Get the relevant links clean and fast without being tracked. - Customization: As it&#x27;s open-source, you can fork it and re-style it any way you want.<p>You can get started with MiniSearch by cloning the repository from GitHub (<a href="https:&#x2F;&#x2F;github.com&#x2F;felladrin&#x2F;MiniSearch">https:&#x2F;&#x2F;github.com&#x2F;felladrin&#x2F;MiniSearch</a>) and running it locally or by using it online on this HugginFace Space: <a href="https:&#x2F;&#x2F;felladrin-minisearch.hf.space" rel="nofollow noreferrer">https:&#x2F;&#x2F;felladrin-minisearch.hf.space</a> (Alternative Space address: <a href="https:&#x2F;&#x2F;huggingface.co&#x2F;spaces&#x2F;Felladrin&#x2F;MiniSearch" rel="nofollow noreferrer">https:&#x2F;&#x2F;huggingface.co&#x2F;spaces&#x2F;Felladrin&#x2F;MiniSearch</a>)<p>You can even set it as your browser&#x27;s address-bar search engine using the query pattern `<a href="https:&#x2F;&#x2F;felladrin-minisearch.hf.space&#x2F;?q=%s" rel="nofollow noreferrer">https:&#x2F;&#x2F;felladrin-minisearch.hf.space&#x2F;?q=%s</a>` (where your query replaces %s).<p>At the moment of this writing, the app is using TinyLlama and LaMini-Flan-T5 models, but there&#x27;s an option to try to use larger models like Mistral 7B (not recommended, though, as it could be slow and break the fast-search experience).<p>That&#x27;s what I had to share. Thanks for reading!<p>Your feedback means the world to me! Please don&#x27;t hesitate to reach out if you have any questions or suggestions or want to learn more.

2 条评论

hackideiomat超过 1 年前
To anyone trying this, the first search is slow. After the first load, it gets much better!<p>&gt; - Privacy: Besides the queries that go anonymously to the actual search engine, nothing else leaves your device.<p>This is one of the main problems I have with my current search engine. The AI features partly use APIs from companies I really don&#x27;t trust with my data (e.g., OpenAI).<p>Regarding the actual search engine: the server appears to use DuckDuckGo on the backend, which would be Bing. If I&#x27;d run this, I&#x27;d probably try to point it at a searx-ng instance or something.
评论 #37888765 未加载
lygten超过 1 年前
Awesome project. I have it installed using Docker. Is there a way of having just show the links, instead of waiting for the details to load. Seems that this would make it much faster. Is there an option to choose an alternative search engine. Edit: Ok. Found it in the code. `<a href="https:&#x2F;&#x2F;links.duckduckgo.com&#x2F;d.js?q=${query}&amp;o=json&amp;vqd=${vqd}" rel="nofollow noreferrer">https:&#x2F;&#x2F;links.duckduckgo.com&#x2F;d.js?q=${query}&amp;o=json&amp;vqd=${vq...</a>`
评论 #37889446 未加载