FreeAskInternet is a free and open-source tool that combines multiple advanced language models to provide search and answer generation services.

We can directly install and run it locally; it is more like a search aggregator for various large language models.

According to the open-source description, FreeAskInternet uses large language models (LLM) to generate answers without relying on a graphics processing unit (GPU). After a user asks a question, the system uses SearXNG to perform multi-engine search, combines the search results with the ChatGPT3.5 language model, and generates answers based on these search results. All processes run locally and do not require a GPU or API keys from OpenAI or Google.

The homepage is still very clean:

AI chat interface based on search:

Supports multiple large language models and custom large language models (such as ollama):

To sum up, FreeAskInternet has the following features:

• Free and open source, we can install it locally by ourselves.
• It uses APIs from ChatGPT3.5, Qwen, Kimi, Zhipu AI, etc. Users do not need to have API keys for these to use them.
• Because it uses free APIs, unlike large language models that usually require a GPU to run, FreeAskInternet can run without a GPU, reducing the hardware requirements for users.
• We can also customize the large language model, such as ollama (download address: https://ollama.com/), which provides users with more flexibility and personalization options.
• Installation is simple. Through Docker Compose, a popular container orchestration tool, users can quickly deploy FreeAskInternet, simplifying the installation and configuration process.

How FreeAskInternet works:For the question entered by the user in the interface, SearXNG performs a multi-engine search. After the search is complete, the system automatically crawls and collects the link content in the search results, and then passes this content to the selected LLM (Large Language Models), such as ChatGPT3.5, Kimi, Qwen, ZhipuAI, or a custom ollama. These models use the passed content as a reference to generate content and answer the user's question.

Installation and Usage

This project is in its early stages, so there may inevitably be some issues, but it is completely sufficient for personal learning and use.

Before installation, you need to have Docker installed locally and support docker-compose. Installation and usage are as follows:

git clone https://github.com/nashsu/FreeAskInternet.git
cd ./FreeAskInternet
docker-compose up -d 

After installation, you can open the link http://localhost:3000 to start using it.