Your AI coding workspace

GlixoCode

Choose your model. Keep your project, conversations and tools together.

Download Glixo Code Checking latest release ▶ Try in browser
WindowsLinuxBrowserAndroid
GLIXO CODE · ANDROIDExperimental

Looking for the
Android version?

A Glixo UFO beaming up a little Android robot

Run Glixo Code directly on your phone. Add Glixo Android Services for on-device tools and local Gemma models, even without an internet connection.

Android is experimental. You can still pair your desktop for the full Glixo Code experience.

Offline use requires initial setup and a downloaded model.

Get it on Google Play

Use the Google account you use on Google Play.

No obligations. No newsletter. We’ll email a link to verify this Google Play account. Access is not active until setup is complete. Read our privacy notice.

Prefer a direct install?

No closed-test signup needed. Download the APKs directly.

Glixo Code APK Glixo Android Services APK

Checking the latest stable Android packages…

Glixo Code desktop — a live coding-agent session with changed files and commits
See it work

Keep the work moving.
Stay in control.

Change models. Keep the conversation.

Switch to another supported model when you need a different approach. Keep the conversation and project tools in the same Glixo session.

Save a point to return to.

Before a risky edit, save a checkpoint. Review the code changes, then restore the saved version if you want to undo them.

From a task to a reviewable change.

Give it a goal — “harden the webhook retry path with a focused regression test”. Follow the edits and test results in the workspace, then review what changed before you keep it.

A workspace for the whole task.

Start with a project and a concrete job: investigate a bug, change a feature, or automate a repeated step. Keep the files, terminal and useful context close to the conversation.

Zero-install browser coding

Open a project in a browser workspace with files, a terminal, Git, and a live preview.

Bring your choice of AI

Connect a supported provider account or API key, or choose a local model. Change supported models within the same session.

Glixo Flow preview: workflow approvals

Glixo Flow is in suite development; this preview is unavailable in the current stable Glixo Code release.

Memory across sessions

Save useful project information and retrieve it in a later session, without repeating the same explanation.

Works with your models
Sandbox engineering

Prepared tools. Shorter commands.

Prepare toolchains once, then use compact command-line tools for everyday project work. These figures come from one recorded six-task Windows sample; results vary with the machine and workload.

Glixo Code Machines dashboard: fleet status with machines, sandboxes, and direct hosts, plus per-machine hosts and sandbox environments across a Windows workstation and a Linux render node
2.30×
command speed in this sample
1.91s in the sandbox vs 4.41s in PowerShell, across a six-task workload.
72.5%
shorter command text
106 vs 385 command-text tokens in the recorded sample.
What this comparison measures
  • Recorded on Windows via Glixo RPC across six command tasks. The sample measures command execution and command text; full AI-task time and provider billing are outside its scope.
  • Windows PowerShell: 4,406.5 ms · Sandbox: 1,913.5 ms — a 2,493 ms difference in this sample.
  • Five of six tasks were faster (jq ran 6.62× faster); cloc was 63.1% slower.
  • Command-text tokens: 385 → 106, a reduction of 279 (72.5%). This is a recorded engineering example; current-release performance needs its own measurement.

Your server. Encrypted payloads.