v1.0 — Open source & on-prem ready

Run AI on
your hardware.

An open-source, self-hosted AI platform. Chat with files, generate audio, analyze data — bring your own models from Ollama, LM Studio, vLLM, or any local engine. Zero data leaves your network.

Quick Install View Source
Terminal
$ curl -fsSL https://get.local-ai.run/install.sh | bash

# ✔ Docker is installed and running
# ✔ Pulling images from Docker Hub...
# ✔ Starting Local AI services...

🟢 Local AI is ready → http://local-ai.localhost

$
// capabilities

Everything runs locally.

Modular AI tools that work offline. Pick what you need, choose your model engine for each feature. Swap providers without changing a line of code.

💬

Chat with Files

Upload PDFs, docs, spreadsheets, and code files. Ask questions, get summaries, extract insights — all processed through local embeddings and RAG.

Any LLM + Vector DB
🔊

Text to Audio Coming Soon

Convert any text into natural-sounding speech. Run TTS models locally with multiple voices and languages. Perfect for accessibility and content creation.

Pluggable TTS Engine
🖼️

Image Generation Coming Soon

Generate images from text prompts using local diffusion models. No API keys, no rate limits, no data leaving your machine.

Stable Diffusion / ComfyUI
📝

Document Summarizer Coming Soon

Drop in lengthy reports, articles, or research papers. Get concise summaries with key takeaways, structured by sections.

Your Choice of LLM
🔍

Semantic Search Coming Soon

Index your files and search by meaning, not keywords. Find related documents, surface connections, and navigate your knowledge base naturally.

Embeddings + Vector DB
🔌

Plugin System Coming Soon

Extend local-ai with community plugins. Build custom tools using the plugin SDK and connect them to any local model or API.

Extensible SDK
// architecture

How it works under the hood.

A containerized stack that runs entirely on your infrastructure. Each service is isolated and independently scalable.

01

Web UI

React frontend served via Nginx

→
02

API Gateway

Django backend orchestrating services

→
03

Model Engine

Ollama · LM Studio · vLLM · any provider

→
04

Storage Layer

Vector DB + file store for RAG & data

// get started

Deploy in under 5 minutes.

Choose your deployment method. All options pull the same pre-built Docker images and produce the same result.

🐳 Docker Compose

Clone the repository and bring up the stack manually. Full control over configuration, environment variables, and service profiles.

git clone https://github.com/360solutions-dev/local-ai
cd local-ai
cp .env.example .env
docker compose up -d
Full control Custom config Advanced users

☸️ Helm Chart Coming Soon

Deploy to Kubernetes for teams and enterprise. Includes resource limits, ingress configuration, persistent volumes, and horizontal pod autoscaling.

helm repo add local-ai https://charts.local-ai.run
helm install local-ai local-ai/local-ai
K8s native Scalable Teams / Enterprise
Deployment Ease of Use Custom Config Auto Updates Multi-node Best For
Quick Install ●●●●● ●●●●● ✓ — Individual devs
Docker Compose ●●●●● ●●●●● — — Power users
Helm Chart Soon ●●●●● ●●●●● ✓ ✓ Teams & Enterprise

Your data. Your models.
Your rules.

Join the community building private, self-hosted AI infrastructure.

Install local-ai ⭐ Star on GitHub