Local-first // Windows + Linux // Apache-2.0

Your open-source AI coworker.
Your rules.

Plan the outcome, bring your own model, connect the tools you need, and keep important actions visible. A local-first alternative for people who want more control than a closed AI workspace can offer.

€0 app price Local by default You choose the model
01 / OLLAMA 02 / MODEL CHOICE 03 / LOCAL FILES 04 / MCP TOOLS 05 / APPROVALS
LocalAI Cowork desktop app with a local Ollama model ready for a task
Real product interfaceDark mode · Ollama · Windows + Linux
No account requiredStandalone desktop starts without a server account.
No website trackingNo analytics, advertising scripts, or tracking cookies.
Inspectable releasesSource, SBOM, checksums, and build attestations.
Windows + LinuxNSIS, AppImage, DEB, and RPM release packages.

01 // The workspace

More than another chat window.

LocalAI Cowork keeps the objective, context, execution, and review surface together—so the work remains understandable while it moves.

01

Outcome first

Work from a visible plan, not a disappearing prompt.

Describe what done looks like. Keep steps, task state, tool activity, and the result connected.

02

Model choice

Use the model that fits the work.

Run locally with Ollama or configure a compatible hosted endpoint when a task needs it.

Ollama OpenAI-compatible OpenRouter
03

Selected context

Your files enter the task deliberately.

Choose the folders, documents, and project instructions that belong to the current objective.

04

Real tools

Connect action to oversight.

Use MCP servers, local tools, terminal workflows, skills, and reusable pipelines without hiding what is happening.

02 // The working loop

Brief. Run. Review. Reuse.

A simple loop turns a request into controlled work and a result you can inspect before it leaves your desk.

01

Set the outcome

Define the result, project, model, and relevant context.

02

Watch the run

Follow the plan, tool calls, progress, and generated artifacts.

03

Approve the risk

Keep sensitive actions behind explicit rules and approval points.

04

Keep what works

Turn repeatable instructions into skills, templates, and pipelines.

03 // Optional distributed mode

Close the laptop.
Keep the Run.

Self-host one Linux control plane when work must continue in the background. The desktop remains fully standalone and private files stay local unless you explicitly create a remote snapshot.

01 / DESKTOP

Local-first, even offline.

Keep projects, Ollama or vLLM endpoints, native tools and durable local Runs on your own Windows or Linux device.

02 / SERVER

One self-hosted HTTPS origin.

Caddy, Rust services, PostgreSQL, encrypted object storage and hardened Linux sandboxes run behind one proxy port.

03 / WEB + ANDROID

Review and control from anywhere.

Follow events, answer questions, approve actions, upload attachments and take over authenticated GUI sessions.

04 / WINDOWS POOL

Real Office when you provide it.

Dedicated outbound-only Windows executors add installed Word, Excel, PowerPoint and Windows desktop automation.

Trusted pilot status

The distributed stack has extensive automated acceptance, but physical Android, Office, identity-provider and reverse-proxy validation remains operator-specific. Review the status matrix before public multi-tenant use.

Setup guide Implementation status

04 // Local by design

Your workspace should stay yours.

The desktop app runs on your device. You choose whether a task uses a local model or a configured remote service, and which files and tools it can reach.

Read the privacy boundaries
CONTROL LEDGERSTATUS
Execution surfaceLOCAL DESKTOP
Supported credentialsOS VAULT
Website analyticsNONE
Telemetry senderNOT IMPLEMENTED
Remote data routeYOUR CONFIGURATION
Risky actionsPOLICY + APPROVAL

05 // Know the trade-offs

An open route to AI cowork.

Claude Cowork, Microsoft Copilot Cowork, and LocalAI Cowork serve different ecosystems. Compare the practical differences without pretending the products are identical.

Built in public

Make the workspace yours.

Download the app, inspect every layer, and help shape an open-source alternative for agentic work.

06 // Before you install

Good questions.
Clear answers.

LocalAI Cowork is early-stage software. These are the boundaries worth knowing before you use it for real work.

What is LocalAI Cowork?

LocalAI Cowork is an open-source Windows and Linux desktop workspace for AI-assisted work. It combines chat, tasks, file context, model configuration, MCP servers, tools, approvals, and audit information in one application.

Is LocalAI Cowork the same as OpenCowork?

No. OpenCowork and Open Cowork are names used by other independent projects. LocalAI Cowork is separate; this OpenCowork alternative guide explains how to compare open cowork tools without mixing up product names.

Is LocalAI Cowork part of the LocalAI project?

No. LocalAI Cowork is an independent project and currently uses Ollama for local model execution. It is not affiliated with or endorsed by the separate LocalAI project at localai.io, Anthropic, or Microsoft.

Can it use a local AI model?

Yes. Ollama is supported for local model execution. You can also configure OpenAI-compatible or OpenRouter endpoints when a hosted model fits the task better.

Does my data ever leave the device?

The app itself runs locally and the current release has no implemented telemetry sender. Requests leave the device only when you choose a remote model endpoint, connector, MCP server, or web-enabled tool. Those services receive the information needed for the action you requested.

Is LocalAI Cowork free?

Yes. The application is available under the Apache License 2.0. A remote model provider may charge for its own API usage; local Ollama use does not require a provider subscription.

Which operating systems are supported?

Windows 10 and 11 are the most mature targets. Linux x86_64 is released as a beta-quality AppImage for broad distribution compatibility, plus native DEB and RPM packages. ARM64, musl-only systems, and Windows-specific sandbox, credential-vault, and Office features are not supported on Linux yet. See the Linux installation guide.