# Needle 2

> 14MB on-device tool-calling model

**Category:** AI Agents  
**Pricing:** open-source  
**URL:** https://github.com/cactus-compute/needle  
**GitHub:** https://github.com/cactus-compute/needle  
**Added:** 2026-08-19

---

## The Hook

Needle 2 is a 14MB model that runs tool-calling on-device.

## What's Good

Cactus Compute built a tool-calling model that fits in 14MB. Runs entirely on-device—no API calls, no latency, no costs per call. Purpose-built for structured function calling, not general chat. The small footprint makes it embeddable anywhere.

## The Catch

Not a chat LLM. Needle does one thing: parse user intent into tool calls. You still need the tools and the orchestration around it. Do not expect conversation or reasoning.

## Verdict

Tiny on-device tool-caller. One job, done locally.