Home Marketplace Docs GitHub Download
Free & Open Source

Chitty Workspace

Your Machine | Your AI | Your Rules

Open Source — Build, Deploy, Automate.
Local-first AI assistant with agents, tools, browser, and marketplace. Bring your own API keys. No cloud server required.

Windows macOS Linux

What is Chitty Workspace?

Everything you need to build with AI

A desktop AI assistant that lives on your machine. Create custom agents with their own tools and instructions. Control your browser to automate any website. Install packages from the marketplace or build your own. Schedule agents to run tasks while you sleep. Use any LLM provider or run models locally on your GPU. Your data never leaves your computer.

Agents

AI-powered agents with custom instructions, tools, and memory. Build specialized assistants for coding, research, automation, and more. Agents can delegate tasks to other agents across panels.

Browser Control

The Chitty Browser Extension gives agents full control of your real Chrome browser. Click, type, navigate, take screenshots, and run JavaScript — using your existing login sessions.

Marketplace

Install tool packages for web scraping, Google Cloud, social media, and more. One-click installs, or build your own packages and share them with the community.

Bring Your Own Keys

Use your own API keys from OpenAI, Anthropic, xAI, or Google. Or run open-source models locally with Ollama and HuggingFace. Your keys, your models, your control.

Supported providers

BYOK & BYOM

Your Keys. Your Models.

Bring Your Own Keys. Bring Your Own Models. No middleman, no lock-in.

BYOK — BRING YOUR OWN KEYS
OpenAI
Cloud · BYOK
Supported Models
o3 · o4-mini · GPT-4.1
GPT-4o · GPT-4o-mini
Get API key →
Anthropic
Cloud · BYOK
Supported Models
Claude Opus 4 · Sonnet 4.6
Claude Sonnet 4 · Haiku 4.5
Get API key →
xAI
Cloud · BYOK
Supported Models
Grok 4 · Grok 4.1 Fast
Auto-discovered from API
Get API key →
Google AI
Cloud · BYOK
Supported Models
Gemini 3 Flash · 2.5 Pro
Gemini 2.5 Flash · 3.1 Lite
Get API key →
BYOM — BRING YOUR OWN MODELS
Ollama
Local Runtime · Free
Open-Source Models
DeepSeek R1 · Llama 3.3
Qwen 3 · Mistral · Gemma 2
GPU Acceleration NVIDIA (CUDA) · AMD (ROCm)
Apple Silicon (Metal) · CPU
Download Ollama →
HuggingFace
GGUF Sidecar · Free
GGUF Model Runtime
Run any quantized GGUF model
via Python inference sidecar
GPU Acceleration NVIDIA (CUDA) · AMD (ROCm)
Apple Silicon (Metal) · CPU
Browse models →
Fine-Tuned Models
Custom · Local
Your Custom Models
Fine-tune and run your own
models locally with full control
Supported Formats GGUF · Ollama Modelfile
LoRA adapters · Custom merges
Learn more →
🔒 Your models. Your machine. Zero data leaks.
✓ One command install
ollama pull llama3 — done.
✓ 100% offline
No API keys, no cloud, no bills.
✓ GPU accelerated
NVIDIA, AMD, Apple Silicon — auto-detected.
✓ Fine-tune & merge
Run your own LoRA adapters and custom models.
Learn how to run local models → Works on any machine — even without a GPU

How it works

Up and running in minutes

Three steps to your own AI-powered workspace.

1

Download & add your API key

Install Chitty Workspace and enter your API key from any supported provider. Keys stay on your machine.

2

Create or pick an agent

Use the Agent Builder to create a custom agent, or start with one of the built-in defaults.

3

Start building

Your agent writes code, runs tools, opens the browser, and automates tasks. You stay in control.

Free forever. No account required.

Download Chitty Workspace and start building with AI today.

Download for free Read the docs

Built by DataVisions