Secure outbound tunnels. Zero inbound ports.

Your local APIs,
available anywhere.

Give Codex, your apps, and your team secure access to LM Studio and private services running on your own hardware—without opening your firewall.

Create a free account See how it works

256-bit token entropy Outbound-only connections In-memory relay

Your machineLM Studio Online

SOCKETSPARKConnect Relay

⌘

AnywhereCodex

{ }

AnywhereYour app

● ● ●

$ curl connect.socketspark.com/v1/models
→ 200 OK · private model ready

NO PORT FORWARDINGOPENAI-COMPATIBLEREVOCABLE PATSRAW TCP TUNNELS

Three steps

From localhost to everywhere

The connector makes one authenticated outbound WebSocket connection. SocketSpark handles the hard parts in between.

✦

Create two PATs

Keep publisher and consumer credentials separate. Revoke either one instantly from your dashboard.

↗

Run the connector

Publish LM Studio or any allowlisted TCP endpoint. Nothing listens on your public network.

⌁

Use your API

Point OpenAI-compatible tools at SocketSpark Connect and authenticate with the consumer PAT.

Built for work in motion

Your compute stays home.
Your workflow doesn’t.

Use the GPU workstation in your office from a laptop at a café. Connect an automation runner to a private development API. Reach a database without exposing it to the internet.

OpenAI-compatible LM Studio endpoints
Named raw TCP endpoints
Independent, expiring credentials
No tunneled payload persistence

● ● ●openai-compatible.http

# Your private model, from anywhere
POST https://connect.socketspark.com/v1/chat/completions
Authorization: Bearer ssc_pat_••••••••
Content-Type: application/json

{ "model": "your-local-model", … }

# The consumer PAT never touches LM Studio

Ready when you are

Make localhost travel.

Create your account

Your local APIs,available anywhere.