# cactus-compute/needle

> A 2-bit, 8-29 MB foundation model for phones, wearables, robots and microcontrollers that handles tool calls, structured extraction and text embeddings on-device.

- Magnitude: 3.0 out of 10 — Steady
- Stars: 13,245 total · +28 stars measured 2026-10-05, 00:08–11:02 UTC, ≈ 67 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: Python · License: Apache-2.0 · Created: 2026-02-24 · Last push: 2026-10-04
- GitHub: https://github.com/cactus-compute/needle · Homepage: https://cactuscompute.com · Page: https://gitnova.dev/en/r/cactus-compute/needle

## Useful for

- Route a voice command to the right app function on a phone
- Extract typed fields from invoices or bookings into a JSON record
- Fine-tune an 8-layer subnetwork on a product's own tools

## Why it’s here

- Star-counter measurements on 2026-10-05 (UTC), 00:08–11:02: 13217 → 13245 stars (+28). This is the change over that interval.
- Estimated end-of-day forecast: about +67 stars, using observed gains and the previous day.
- Hacker News: “Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash” — 236 points, 17 days ago.
- Recent forks include notable developers: @vaimalaviya1233 (576 followers).

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 894
- Issues and pull requests: 165
- Watchers: 73
- Average over the last week: 70 per day
- Usual pace: 123 per day

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-10-05 … 2026-11-03: 11, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None

## Magnitude by day, last 14 days

09-22 3.8, 09-23 4.1, 09-24 4.0, 09-25 3.4, 09-26 3.0, 09-27 2.9, 09-28 2.5, 09-29 2.4, 09-30 2.2, 10-01 2.0, 10-02 2.1, 10-03 2.8, 10-04 3.4, 10-05 3.1

## Hacker News

- Whistle: Speech to Text in 16.9 MB — 962 points, 185 comments: https://news.ycombinator.com/item?id=50008427
- Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash — 236 points, 92 comments: https://news.ycombinator.com/item?id=49748553
- Whistle: Speech to Text in 16.9 MB — 6 points, 1 comments: https://news.ycombinator.com/item?id=49942095

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-05 11:14 UTC, updated every 30 minutes.