EN | δΈ­ζ–‡

Share Your Stable Diffusion Rig. Keep Your Art Yours.

Run image generation on your own GPU with your own models and workflows. Let collaborators access the service remotely without public web exposure or a third-party relay in the data path.

⬇ Download Free
πŸ›‘ Privacy — End-to-end encrypted. No MTCode infrastructure in your data path.
⚑ Simplicity — No public IP, domain, VPN, or port forwarding.
πŸš€ Performance — Direct peer-to-peer data path, with no relay hop.

The Problem

Artists and creative teams often prefer private image-generation workflows for control over models, prompts, outputs, and style exploration. Cloud tools can be convenient, but they may impose content rules, limit custom checkpoints, or handle prompts and outputs under provider-specific policies.

Self-hosted tools such as Stable Diffusion WebUI, InvokeAI, ComfyUI, and custom LoRA or DreamBooth workflows solve the control problem, but they introduce an access problem. The GPU computer sits in your studio or home, while collaborators, clients, or teammates may be elsewhere.

Sharing a public web UI is risky. SSH is too much access. Tunnel services relay traffic through third-party infrastructure. MTCode DirectLink lets you share the image-generation service itself while keeping the computer private.

What MTCode Does

🎨

Publish your existing workflow

Register the local port for Stable Diffusion WebUI, ComfyUI, InvokeAI, MTCode-Diffusion, or another image-generation service. Users access it through MTCode Portal as if it were local.

πŸ”—

Collaborate without public exposure

Invite collaborators to the specific service. They do not need SSH access, router access, or a public URL for your studio machine.

🧠

Keep your model stack

Use your own checkpoints, LoRAs, ControlNet workflows, prompts, and output directories on hardware you control.

How It Works in This Scenario

  1. Run your image-generation server on your GPU computer.
  2. Install MTCode Server and register the service port.
  3. Invite collaborators, clients, or teammates who should access the service.
  4. They install MTCode Portal, sign in, and connect to the service from their own computer.
  5. Application data flows directly between Portal and your server over an encrypted peer-to-peer channel.

Why This Matters

Getting Started

  1. Start your image-generation service locally.
  2. Publish it with MTCode Server.
  3. Invite collaborators and have them connect through MTCode Portal.
Publish a Service   Download