Claude Skill
RightNow-AI/picolm
Open-source C project for running billion-parameter LLMs on low-cost embedded hardware. Enables AI inference on $10 boards with only 256MB RAM. Supports ARM, RISC-V.
Overview
Repository
Install this Skill
git clone https://github.com/rightnow-ai/picolm.gitRegistry
Summary
Picolm is a C-based project that enables running a 1-billion parameter large language model on low-cost embedded hardware with only 256MB of RAM, such as a $10 development board.
在256MB内存的10美元开发板上运行10亿参数大语言模型
Key features
- Runs 1B parameter LLM on embedded hardware
- Minimal memory requirement (256MB RAM)
- Optimized for low-cost ($10) development boards
- Written in efficient C language
- Supports ARM and RISC-V architectures
Use cases
- On-device AI inference for IoT devices
- Educational projects on resource-constrained hardware
- Prototyping LLM applications on Raspberry Pi
- Edge computing with large language models
- Demonstrating efficient model quantization techniques
README excerpt
<p align="center"> <img src="https://img.shields.io/badge/Language-C11-blue?style=flat-square" alt="C11"> <img src="https://img.shields.io/badge/Binary_Size-~80KB-brightgreen?style=flat-square" alt="Binary Size"> <img src="https://img.shields.io/badge/Runtime_RAM-45MB-orange?style=flat-square" alt="RAM"> <img src="https://img.shields.io/badge/Dependencies-Zero-success?style=flat-square" alt="Zero Dependencies"> <img src="https://img.shields.io/badge/License-MIT-yellow?style=flat-square" alt="MIT License"> </p> <h1 align="center">PicoLM</h1> <p align="center"> <strong>Run a 1-billion parameter LLM on a $10 board with 256MB RAM.</strong><br> Pure C. Zero dependencies. One binary. No Python. No cloud. </p> <p align="center"> <code>echo "Explain gravity" | ./picolm model.gguf -n 100 -j 4</code> </p> --- ## The Perfect Match: PicoLM + PicoClaw <div align="center"> <img src="picolm.jpg" alt="PicoLM — Run a 1-billion parameter LLM on a $10 board" width="640"> <br><br> </div> PicoLM was built as the **local brain** for [PicoClaw](https://github.com/sipeed/picoclaw) — an ultra-lightweight AI assistant in Go that runs on $10 hardware. Together, they form a **fully offline AI agent** — no cloud, no API keys, no internet, no monthly bills. > **Every other LLM provider needs the internet. PicoLM doesn't.** <table align="center"> <tr align="center"> <td><b>The Hardware</b></td> <td><b>The Architecture</b></td> </tr> <tr> <td align="center"><img src="https://raw.githubusercontent.com/sipeed/picoclaw/main/assets/licheervnano.png" alt="$9.90 LicheeRV Nano" width="360"></td> <td align="center"><img src="https://raw.githubusercontent.com/sipeed/picoclaw/main/assets/arch.jpg" alt="PicoClaw architecture — PicoLM sits in the LLM box" width="