---
format: "aidr-story-markdown/v1"
id: "d7dddac5c9c6043e7c069229478b43302339e3f52166dba9a681e74708c0c1d2"
canonical_url: "https://aidr.today/d7dddac5?lang=en"
title: "Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-09-18T00:11:44.000Z"
category: "Models"
topics: ["cactus-needle-3","mobile-ai","edge-computing","tool-call","deepseek"]
source_urls: ["https://cactuscompute.com/needle","https://news.ycombinator.com/item?id=49748553"]
summary: "A foundation model for mobile, wearables, robots, smart home, automotive and microcontrollers. One 8-29 MB binary that beats models 10x its size on mobile tool calls and matches 2-3x bigger models on extraction."
---

# Show HN: Cactus Needle 3: 8\-29MB automation models can match DeepSeek V4 Flash

> [Open the canonical story](<https://aidr.today/d7dddac5?lang=en>)

**Published:** 2026-09-18T00:11:44.000Z
**Category:** Models
**Topics:** cactus\-needle\-3, mobile\-ai, edge\-computing, tool\-call, deepseek

## Summary

A foundation model for mobile, wearables, robots, smart home, automotive and microcontrollers\. One 8\-29 MB binary that beats models 10x its size on mobile tool calls and matches 2\-3x bigger models on extraction\.

## Sources

- [Story source](<https://cactuscompute.com/needle>)
- [Discussion](<https://news.ycombinator.com/item?id=49748553>)