Perplexica is a self-hosted, privacy-focused AI answering engine that runs entirely on your own hardware. It combines web knowledge with support for local LLMs (Ollama) and cloud providers (OpenAI, Claude, Google Gemini, Groq), delivering accurate answers with cited sources while keeping searches completely private.
Perplexica positions itself as an open-source alternative to hosted AI search tools. The project emphasizes running on your own infrastructure so teams can keep data, prompts, and logs within their security boundary. The name “Perplexica” draws inspiration from AI-powered answering engines like Perplexity AI, but focuses on self-hosted deployments for privacy-conscious users.
The project follows key principles common in the GenAI ecosystem:
Like many GenAI tools, Perplexica evolved from a simple UI around model access into a comprehensive platform with:
Perplexica uses a bundled approach for ease of deployment:
The Docker image bundles SearxNG internally, making it easy to run a full stack with a single command. A slim version is available for users who prefer to run SearxNG separately.
| Metric | Value |
|---|---|
| Latest Release | v1.12.1 (December 31, 2025) |
| GitHub Stars | 32.2k+ |
| Forks | 3.5k+ |
| Primary Language | TypeScript (98.7%) |
| License | MIT |