Stitch: Open Source Local AI Workspace
Stitch is an open-source, locally-run AI desktop app that integrates memory, meeting recordings, task management, automation, email, web browsing, and more. It works with multiple AI providers or local models, requires no accounts, and keeps all data on your machine for maximum privacy.
Stitch
We're building managed services so you don't have to bring your own API keys or run anything locally. Enter your email and select what interests you.
You're on the list. We'll be in touch.
Capabilities
Everything you need, built in
Stitch comes packed with tools to handle your daily work, from managing your inbox to recording meetings and automating repetitive tasks.
Memory
Stitch remembers your preferences, workflows, and key facts across conversations. It learns what matters to you and applies that context automatically, no repetition needed.
Meeting recordings
Record meetings from Zoom, Google Meet, Teams, Slack, or Discord. Stitch captures audio locally, transcribes it, and generates summaries with action items, all on your machine.
Agenda & todos
A built-in task manager with priorities, due dates, and multiple lists. Stitch can create tasks from your conversations and meetings, so nothing slips through the cracks.
Daily automations
Turn any chat into a recurring workflow. "Summarize my emails every morning" or "clean up my downloads every Friday." Stitch runs it on schedule without being asked.
Email & calendar
Connect Gmail and Google Calendar. Stitch can read, draft, and send emails, manage your schedule, and handle routine communications.
Web browsing
Research topics, look up information, fill out forms, and pull data from websites. Stitch drives a real browser on your behalf.
MCP servers
Extend Stitch with any Model Context Protocol server. Add tools, data sources, and integrations from the growing MCP ecosystem — no plugin development needed.
Skills
Install community-built skills that teach Stitch new capabilities — from code review to data analysis. Plug in expertise on demand.
More coming soon.
How it works
Download, configure, go
Get started in minutes. Stitch handles the complexity so you can focus on what you want to accomplish.
1
Install & connect a provider
Download Stitch and connect your AI provider (OpenAI, Anthropic, Google, or a local model). Bring your own API keys.
2
Describe your task
Tell Stitch what you need in plain language. Organize files, draft an email, research a topic, clean up your schedule. Anything you'd do yourself at a computer.
3
Review & approve
Stitch asks for permission before running sensitive actions. You stay in control while the AI handles the work.
Installation
Getting started on your platform
Stitch is available for macOS and Windows. Your operating system may show a warning the first time you open it. Here's how to get past it.
macOS
Download the .dmg file and open it
Drag Stitch into your Applications folder
Open Stitch from your Applications folder
Windows
Download the .exe installer and run it
Windows SmartScreen may show a "Windows protected your PC" warning
Click "More info"
Click "Run anyway"
Follow the installer prompts to finish the setup
The warning appears because the Windows build isn't code-signed yet. We are working on it. Stitch is fully open source. You can inspect the code or build from source on GitHub.
Extensibility
Bring your own models
Stitch is not locked to a single provider. Use the models and integrations that work for you.
OpenAI
Anthropic
Google Gemini
Local models
Get started for free
Free API keys for LLM & Speech-to-Text
You don't need to spend a cent to use Stitch. These providers offer free tiers with generous limits — no credit card required.
Free LLM providers
Go to aistudio.google.com/live and sign in with your Google account
Click the API key button, which looks like a key
Create a Google Cloud project, or import one if you already have it
Click Create API key
Copy the generated Gemini API key
Paste the key into Stitch settings under the Google Gemini provider
Google AI Studio offers free Gemini API access with rate limits. No credit card is required to generate an API key. Usage is subject to Google's Gemini API terms and free tier limits.
Go to openrouter.ai and sign up with email, GitHub, or Google
Open the top-right dropdown, then go to Preferences → API Keys in the sidebar
Click New Key, give it a name, select No Expiration, and create it
Copy the key — it's only shown once
Paste the key into Stitch settings under the OpenRouter provider
Free tier includes 28+ models (DeepSeek R1, Gemma 3, Gemini Flash, and more) at 20 requests/min. No credit card needed. Use the :free model variant for zero-cost access.
Go to build.nvidia.com and sign up for the free NVIDIA Developer Program
Open your profile from the top-right menu and click API Keys
Click Generate API Key, give it a name, and select Never Expire
Copy your API key (starts with nvapi-) from the popup
Paste the key into Stitch settings under the NVIDIA provider
Free tier gives you 1,000 inference credits on signup (up to 5,000 on request) with a 40 requests/min rate limit. Includes models like Nemotron, Llama, and DeepSeek. No credit card required. OpenAI-compatible endpoint.
Free Speech-to-Text provider
Go to elevenlabs.io and create a free account
Click Developers in the left sidebar
Select the API Keys tab
Click Create API Key, give it a name, and enable Speech-to-Text access
Copy the key immediately — it's only shown in full once
Paste the key into Stitch settings under the ElevenLabs STT provider
The free plan includes 10,000 credits/month which covers both Text-to-Speech and Speech-to-Text (Scribe). No credit card required. Used for transcribing meeting recordings in Stitch.
These providers are independent third-party services and are not affiliated with or endorsed by Stitch. Each provider has its own terms of service, privacy policy, and data handling practices. Please review the respective provider's website for details on how your data is used.
Privacy & trust
No middleman between you and your data
Stitch runs entirely on your computer. There are no Stitch servers, no accounts, and no tracking. When you use a cloud AI provider, your prompts are sent directly to that provider's API. For fully offline use, connect a local model instead.
Runs on your machine
Stitch itself is a desktop app with no backend. Your files, recordings, memories, and todos are stored locally. Nothing is ever uploaded to Stitch servers. There are none.
No accounts or tracking
No sign-ups, no telemetry, no analytics. Download the app and start using it immediately. Stitch doesn't know who you are and doesn't want to.
Direct-to-provider AI
When you use OpenAI, Anthropic, or Google, your prompts go straight from your machine to their API, not through us. Review each provider's data policies to understand how they handle your requests.
Fully offline option
Connect a local model and nothing leaves your computer at all. No internet connection needed, no data shared with anyone. Complete privacy for sensitive work.
Fully open source
Every line of code is public on GitHub. Audit what it does, build it from source, or contribute. No black boxes, no proprietary backends, no trust required.
You approve every action
Stitch asks for your permission before doing anything sensitive. You see exactly what it plans to do and can say no. You're always in the loop.
Try Stitch. It's free and private.
Manage meetings, automate tasks, and let AI remember the details. Open source, runs locally, no account required.
Windows
macOS
GitHub