FROM FIRST IDEA TO NEXT COMMIT

Your terminal.Your models.Get building.

An AI coding agent built for real projects. Build, debug, and move work forward with local or cloud models, in the terminal you already know.

Get AX Code See the workflow

Standard · Open source and free

ax-code — ~/your-next-project

___________  __     _________________________________
___    |_  |/ /     __  ____/_  __ \__  __ \__  ____/
__  /| |_    /_______  /    _  / / /_  / / /_  __/   
_  ___ |    |_/_____/ /___  / /_/ /_  /_/ /_  /___   
/_/  |_/_/|_|       \____/  \____/ /_____/ /_____/

Find the bug. Fix it. Show me what changed.

✓Inspecting the projectsrc/

✓Preparing a focused change+8 −3

✓Running repository checkstest

↳

Ready for your reviewsrc/utils/validate.ts +8 −3

›What are we building next?

AX CodeYour project. Your next move.

src/utils/validate.ts +8 −3

  export function isValid(value) {−   return value.length > 0;+   if (typeof value !== 'string') {+     return false;+   }+   return value.trim().length > 0;  }

Inspect execution records, review file changes, and compare runs before deciding what to keep.

✓Ready for your review

Illustrative workflow01 / AX CODE

macOS / Windows / UbuntuLocal + cloud models13 interface languagesYour existing workflow

01 / LESS FRICTION. MORE BUILDING.

Made for the work between commits.

From understanding an unfamiliar codebase to reviewing a change, keep your development workflow in one place.

Stay in your flow.

Explore files, make changes, and run your project’s tools from a familiar terminal workspace.

Choose your intelligence.

Connect a supported local runtime or cloud provider. Pick the model that fits the task and your hardware.

See what changed.

Inspect execution records, review file changes, and compare runs before deciding what to keep.

02 / LOCAL INFERENCE, REAL PROJECTS

Put your hardware to work.

Bring local models into your everyday development. Keep inference on your machine, with a connected workflow from model selection to code review.

Better together with AX Engine.

A native local inference experience on supported Apple Silicon Macs. AX Code also connects to other supported local runtimes on Windows, Ubuntu, and macOS.

Explore AX Engine ↗

Hardware requirements depend on your model. External providers and connected tools may send data off-device.

YOUR MACHINE

AX CodeCoding workspace

AX EngineLocal inference

Your choice of model

Local model guide ↗

03 / AX CODE BUSINESS · PLANNED

Your team. Your environment.

Bring AI coding into your company’s systems. AX Code Business is being built around AX Trust and the AutomatosX ecosystem, with company-managed AI access and data workflows.

Explore Business

01AX CodeCoding workspace

02AX TrustCompany-managed models

03AutomatosXConnected workflows

Business-controlled records

04 / START YOUR WAY

Personal projects. Shared ambitions.

Standardv8.0

Free / Standard

Open-source AI coding for individuals and teams, with local and cloud models.

Terminal coding workflow
Supported local and cloud models
13 interface languages
Execution records and change review

Get started

AX Code Business

Let’s talk

Planned commercial edition with AX Trust governance and AutomatosX workflows.

Native AX Trust integration
AutomatosX ecosystem workflows
Company-managed data and AI access
Deployment scope agreed with your team

Explore Business

Standard is free for personal and commercial use. v8.0 downloads are upcoming; Business pricing and terms will be announced separately. Model-provider charges and hardware costs may apply.

FAQ

A few things to know.

Where does AX Code run?

AX Code targets macOS on Apple Silicon, Windows x64/ARM64, and Ubuntu 24.04+ amd64/arm64. AX Engine’s native local inference path is specific to supported Apple Silicon Macs. Check the installation guide for current release requirements.

Is AX Code free?

AX Code Standard is open source under Apache-2.0 and free for personal and commercial use. Business is a planned, separately licensed proprietary edition for organization-wide governance and integrations. Model-provider usage may be billed separately.

Can I work with local models?

Yes. Connect a supported local runtime, including AX Engine on eligible Macs. Model fit, speed, and tool support depend on your hardware and the selected model. Cloud models remain an option.

Does my code stay on my computer?

Local inference processes model requests on your machine. Cloud providers, connected tools, and network commands can transmit data. Your chosen configuration determines the data flow; local inference alone does not mean every action is offline.

What is coming in v8.1?

Computer use (CUA) is planned for v8.1, extending development workflows into supported applications. It is not included in the v8.0 launch promise. Availability and platform details will be announced with that release.

On the horizon

v8.0: open-source Standard + proprietary Business direction. v8.1: computer use (CUA), planned.

Get AX Code →

Your next project starts here.

Bring an idea. Bring your models. Make it happen.

Get AX Code