What is FreeAskInternet?
FreeAskInternet is a completely free, private, and locally running search aggregator and answer generator using LLMs. It allows users to ask questions, performs multi-engine searches using searxng, and combines the search results with LLMs like ChatGPT3.5 to generate answers. All processes run locally without needing GPUs or API keys.
Key Features
- Completely FREE (no API keys needed)
- Completely LOCAL (runs on any computer)
- Completely PRIVATE (all processes run locally)
- Runs WITHOUT LLM Hardware (NO GPU needed)
- Uses free ChatGPT3.5 / Qwen / Kimi / ZhipuAI APIs
- Custom LLM support (e.g., Ollama)
- Fast and easy deployment with Docker Compose
- Web and Mobile friendly interface
How It Works
- User inputs question in the FreeAskInternet UI.
- The system performs a multi-engine search using searxng.
- It crawls search result links and uses LLMs to generate an answer based on the contents.
- Streams the answer to the Chat UI.
- Supports custom LLM settings for infinite LLM support.
Running Locally
git clone https://github.com/nashsu/FreeAskInternet.git
cd ./FreeAskInternet
docker-compose up -d
Open the web interface at http://localhost:3000.
Using Custom LLMs
- Start Ollama server:
export OLLAMA_HOST=0.0.0.0 ollama serve
- Set Ollama URL in settings using your computer's IP address.
Credits
- ChatGPT-Next-Web: ChatGPT-Next-Web
- FreeGPT35: FreeGPT35
- Kimi/Qwen/ZhipuAI: LLM-Red-Team
- Searxng: Searxng
License
Apache-2.0 license