Logo
Deploy Now

SaaS Alternative

Perplexity Google Search You.com Bing Copilot

Categories

Self-Hosted Privacy AI LLM Research Search Engine Machine Learning Information Retrieval

Stars

3533

Forks

320

Watchers

32

Developer links

Last Updated: 07-23-2026 03:41 PM (UTC)

Farfalle

Live web search plus an LLM of your choice: Farfalle is an open-source, self-hosted answer engine in the Perplexity mold. Queries route through one of several search providers - self-hosted SearXNG for a fully independent stack, or Tavily, Serper, and Bing APIs - and the model composes a cited answer from the retrieved results. Model flexibility is the core design: run llama3, mistral, gemma, or phi3 locally through Ollama for zero per-query cost and full privacy, use cloud models like GPT-4o or Groq-hosted Llama 3 for speed, or route to any provider via LiteLLM. An Expert Search mode uses an agent that plans a multi-step search strategy and executes it for harder questions, and chat history keeps prior research sessions available. The stack is a Next.js and shadcn/ui frontend over a FastAPI backend with Redis rate limiting, shipped as a pre-built Docker image. A browser search-engine entry pointing at your instance makes it the default search from the address bar. Paired with SearXNG and Ollama, the whole pipeline runs with no external API at all.

Farfalle

Benefits

  • Fully Independent Search Stack
  • With SearXNG for retrieval and Ollama for inference, the entire pipeline runs on your infrastructure - no API keys, no per-query fees, no query logs leaving your server.
  • Choose Speed, Cost, or Privacy per Query
  • Local models for sensitive questions, Groq for fast cheap inference, GPT-4o when quality matters most - all in the same interface via LiteLLM routing.
  • Answers, Not Link Lists
  • Live results are synthesized into a direct, cited response, cutting the tab-opening and skimming that raw search demands.
  • Your Browser's Default Search
  • One search-engine entry pointing at your instance puts AI answers behind the address bar without a subscription.

Features

  • Multiple Search Providers
  • SearXNG for self-hosted metasearch, plus Tavily, Serper, and Bing API options.
  • Local and Cloud LLMs
  • llama3, mistral, gemma, and phi3 via Ollama; GPT-4o and Groq Llama 3 in the cloud; any other provider through LiteLLM.
  • Expert Search Agent
  • An agent plans and executes multi-step search strategies for complex questions instead of a single retrieval pass.
  • Chat History
  • Prior research conversations persist for follow-up and reference.
  • Modern Stack
  • Next.js with shadcn/ui frontend, FastAPI backend, Redis rate limiting, and Logfire logging.
  • One-Container Deployment
  • Pre-built Docker image with Docker Compose setup, configurable through a single .env file.