Neura Market
LLocalSearch logo

LLocalSearch

Free

LLocalSearch is a completely locally running search aggregator using LLM Agents. The user can ask a question and the system will use a chain of LLMs to find the answer. The user can see the progress of the agents and the final answer. No OpenAI or Google API keys are needed. ![GitHub Repo stars](htt

FreeFree tier
Inputs: textOutputs: text
Type
Open Source
Visit Website GitHub (40 stars)

About LLocalSearch

LLocalSearch is a completely locally running search aggregator that uses LLM agents to answer questions. It runs entirely on local hardware without requiring any API keys from OpenAI or Google. The system employs a recursive chain of LLMs that can choose to use tools (like web search) multiple times based on the query. Users can see live progress logs, source links, and the final answer. It supports follow-up questions, is mobile-friendly, and offers dark/light mode. Note: The project was archived by the owner in June 2026 and is no longer under active development; a rewrite/relaunch is planned in private beta.

Key Features

Completely local execution (no API keys required)
Recursive LLM agent system that can use web search tools multiple times
Live progress logs and source links for transparency
Supports follow-up questions
Mobile-friendly design with dark/light mode
Runs on low-end hardware (demo uses 300€ GPU)

Pros & Cons

Pros
  • No API keys needed, enhancing privacy
  • Runs on relatively low-end hardware
  • Live logs and source links allow users to verify answers
  • Follow-up questions are supported
  • Open source and free to use
Cons
  • Archived project, no longer under active development
  • Requires local setup and a capable local LLM (less smart than cloud models)
  • Known issue with LLama3 stop words causing hallucination at end of turn
  • Not suitable for users without local GPU or sufficient hardware

Best For

Private search aggregation without relying on cloud servicesResearch with transparent agent reasoning and source verificationOffline-friendly environment for querying the web via local LLMs

FAQ

Does LLocalSearch require any API keys?
No. It runs completely locally and does not need OpenAI, Google, or any other cloud API keys.
Can it run on low-end hardware?
Yes. The demo video was recorded using a 300€ GPU, and the tool is designed to run on relatively low-end hardware.
Is the project still actively maintained?
No. The repository was archived by the owner on June 1, 2026, and is now read-only. The owner is working on a rewrite/relaunch within a private beta.