Veloresearch

VLS 32M Context Local LLM — AI skill for Claude Code

AI community

Give Claude Code, Codex and opencode a 32M-token memory that never leaves your machine.

How to install VLS 32M Context Local LLM

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Veloresearch/VLS-32M-Context-Local-LLM and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What VLS 32M Context Local LLM does

Give Claude Code, Codex and opencode a 32M-token memory that never leaves your machine. Local LLM server (MCP + OpenAI API) that runs any GGUF model with a context far larger than its window - no cloud, no retraining, no vector database. Windows and Linux, one command.

Alternatives in AI

Optimization — ai-research-skills AWQ, GPTQ, GGUF, Flash Attention, bitsandbytes 5.4k ★
Token Monitor — Local-first desktop widget for tracking token usage, costs, and limits across 32+ AI coding tools—including Cl 1.8k ★
Paritok 4b V1 — Non-destructive compression gateway for AI coding agents 1.4k ★

README

VLS — a model with an 8,192-token window, answering across 32,000,000

[![Release](https://img.shields.io/github/v/release/Veloresearch/VLS-32M-Context-Local-LLM?label=release&color=4B5BF5&labelColor=17171A&style=for-the-badge)](https://github.com/Veloresearch/VLS-32M-Context-Local-LLM/releases/latest) [![Downloads](https://img.shields.io/github/downloads/Veloresearch/VLS-32M-Context-Local-LLM/total?label=downloads&color=2EA043&labelColor=17171A&style=for-the-badge)](https://github.com/Veloresearch/VLS-32M-Context-Local-LLM/releases) [![Stars](https://img.shields.io/github/stars/Veloresearch/VLS-32M-Context-Local-LLM?style=for-the-badge&color=E3B341&labelColor=17171A&logo=github&logoColor=white)](https://github.com/Veloresearch/VLS-32M-Context-Local-LLM) [![Views](https://visitor-badge.laobi.icu/badge?page_id=Veloresearch.VLS-32M-Context-Local-LLM&title=views&color=00A3A3&style=for-the-badge)](https://github.com/Veloresearch/VLS-32M-Context-Local-LLM)

[![Windows](https://img.shields.io/badge/Windows-x64_%C2%B7_CUDA-0078D4?labelColor=17171A&style=for-the-badge&logo=windows&logoColor=white)](#windows) [![Linux](https://img.shields.io/badge/Linux-x64_%C2%B7_Vulkan-E95420?labelColor=17171A&style=for-the-badge&logo=linux&logoColor=white)](#linux) [![Licence](https://img.shields.io/badge/free_for_personal_use-8B5CF6?labelColor=17171A&style=for-the-badge)](LICENSE)

**[Install](#install)** · **[Velocity Context](#velocity-context)** · **[Everything in it](#everything-in-it)** · **[Mesh](#mesh)** · **[Benchmarks](#measured-not-claimed)** · **[API](#connect-anything)** · **[FAQ](#questions-people-actually-ask)**

**[veloresearch.com/vls](https://veloresearch.com/vls)**