nashsu/freeaskinternet

★ 8,735PythonAudience · developerComplexity · 3/5LicenseSetup · moderate

Mindmap

mindmap
  root((FreeAskInternet))
    What it does
      Multi-engine web search
      AI answer generation
      Private and self-hosted
    AI Models
      Free external APIs
      Local Ollama models
      No GPU needed
    Search
      SearXNG aggregator
      Multiple engines
    Setup
      Docker Compose
      Single command start
    Audience
      Privacy-focused users
      Developers

mindmap root((FreeAskInternet)) What it does Multi-engine web search AI answer generation Private and self-hosted AI Models Free external APIs Local Ollama models No GPU needed Search SearXNG aggregator Multiple engines Setup Docker Compose Single command start Audience Privacy-focused users Developers

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Ask research questions and get AI-written answers sourced from live web searches, entirely on your own machine.

USE CASE 2

Replace paid AI search tools like Perplexity with a free, private alternative that requires no accounts or API keys.

USE CASE 3

Run fully offline AI question-answering by connecting the tool to a local model via Ollama.

USE CASE 4

Build a private knowledge search tool on top of the self-hosted stack.

Tech stack

PythonDockerSearXNGOllama

Getting it running

Difficulty · moderate Time to first run · 30min

Requires Docker and Docker Compose, no GPU or API keys needed.

Apache 2.0, free to use, modify, and distribute for any purpose including commercial, as long as you keep the license notice.

In plain English

FreeAskInternet is a self-hosted tool that works similarly to AI-powered search products like Perplexity. You type a question, the system searches the web across multiple search engines at once, pulls the content from those results, feeds that content to an AI language model, and returns a written answer based on what it found. Everything runs on your own computer, and none of your questions or results leave your machine. The key draw is that it requires no API keys and no payment. It uses a local search aggregator called SearXNG to gather results from multiple search engines privately. For the AI step, it connects to freely accessible versions of models like ChatGPT 3.5, as well as Chinese models like Kimi, Qwen, and ZhipuAI, without needing your own account keys. If you prefer to run the AI component fully offline, it also supports local models through Ollama, which lets you run open AI models on your own hardware without any internet connection for that step. Importantly, no GPU is required. The system is designed to run on a normal computer without expensive hardware. The AI processing either uses a free external API call (with no cost to you) or a local model that can run on a CPU. Setup is handled through Docker Compose, which is a tool that starts multiple software components together from a single command. After cloning the repository and running that command, the web interface becomes available at a local address in your browser. It works on both desktop and mobile browsers. The project is in early development, and the README notes that bugs should be expected. It is released under the Apache 2.0 open source license.

Copy-paste prompts

Prompt 1

How do I set up nashsu/freeaskinternet on my computer using Docker Compose? Walk me through every step from cloning the repo to opening the web interface in my browser.

Prompt 2

How do I configure freeaskinternet to use a local Ollama model instead of the free external APIs so my searches are completely offline?

Prompt 3

I want to add a new free AI model to freeaskinternet. Where in the code do I register it and what configuration does it need?

Prompt 4

How does freeaskinternet compare to Perplexity AI? What can it do that Perplexity cannot, and what are the trade-offs?

Open on GitHub → Explain another repo

← nashsu on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.