← Discover
Discover · GitHub · Needle↑ +315 today

Lightweight AI for edge devices

Developed by Cactus Compute, Needle offers a 14 MB foundation model that can run on small hardware such as phones, wearable devices and robots. This lightweight structure aims to run artificial intelligence applications locally on edge devices with limited processing power.

Updates

  • September 7, 2026: Stars 9,378 → 10,436.
  • August 27, 2026: Stars 8,376 → 9,378.
  • August 22, 2026: Stars 7,348 → 8,376.
  • August 18, 2026: Stars 6,022 → 7,348.

What you get

  • Single file model with only 14 MB size
  • Working with 28 MB RAM with low memory usage
  • Ability to summon vehicles and extract structured data

Installation

Basic setup
pip install cactus-needle
Installation with GPU support
pip install "cactus-needle[gpu]"

If you don't write code

🤖 Paste this into your AI agent (Claude Code · Codex · Antigravity)

You are an artificial intelligence assistant. Analyze the text coming from the user and make the necessary tool calls using the Python functions I defined or extract the desired data from the text in structured JSON format. Always consider the confidence score provided by the model in your answers and confirm the transaction at low confidence levels.

Related dictionary terms

Who it is forIt is suitable for developers looking for a fast and lightweight AI model that will run natively on edge devices with limited hardware resources.
LicenseMIT

Links

TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-08-13: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.