Tunneling like never before
Expose your locally running AI models, web applications, and APIs directly to the internet — instantly, securely, and without the usual headache. One command is all it takes.
What you'll be able to do
We're building tunneling for the modern developer — whether you're training machine learning models, iterating on a side project, or demoing to a client.
Host AI Models Locally, Serve Globally
Training a model on your GPU? Expose your local inference endpoint to the internet so teammates, clients, or your own cloud services can hit it directly — no deployment needed.
One Command Setup
No YAML files. No Docker configs. No port forwarding nightmares. Just run one command and your local server gets a public HTTPS URL in under two seconds.
Share Work-in-Progress Instantly
Working on a frontend? An API? A webhook handler? Share a link with your team to get real-time feedback without pushing to staging.
End-to-End Encryption
Every tunnel runs over HTTPS with TLS termination. Your traffic is encrypted from the browser all the way to your machine. We never store or inspect your data.
Webhook Development Made Real
Stop deploying to test webhooks. Point Stripe, GitHub, or Shopify directly at your local server through a tunnel. See requests hit your code in real-time.
Custom Subdomains
Get a predictable, memorable URL for your tunnels. Reserve a subdomain like your-app.requex.me so your integrations never break between sessions.
How it works
Three steps. Zero complexity.
Start your app locally
Run your web server, API, or AI model on any port. It could be a Python Flask server serving model predictions, a Next.js frontend, or a raw HTTP server — it doesn't matter.
Run the tunnel command
A single CLI command — requex tunnel --port 8080 — and we instantly assign a public HTTPS URL. The entire handshake happens in milliseconds through our edge infrastructure.
Share the link, receive traffic
Your public URL is live. Anyone on the internet can hit it, and the requests flow directly to your local machine. Inspect, debug, and iterate — all without deploying anything.
Why we're building this
We started Requex as a simple webhook testing tool — give developers a URL, let them see what's coming in, and help them build faster. But the more we talked to developers, the more we heard the same frustration:
“I just need my localhost on the internet for 10 minutes. Why do I need to sign up, install a binary, configure a YAML, and pay $20/month?”
The numbers speak for themselves. Over 73% of developers working on API integrations need to expose their local environment to external services at some point during development. Yet the most popular tunneling tools haven't evolved in years — they still rate-limit free plans, force session timeouts, and charge premium prices for features that should be baseline.
Meanwhile, the landscape has shifted dramatically. The rise of local-first AI development has created an entirely new use case. Developers are training LLMs, running Stable Diffusion, serving LLaMA models, and building RAG pipelines — all on their own hardware. They need a way to let external services, teammates, or even their own mobile apps reach these local endpoints. Deploying to a cloud GPU just to share a model demo is expensive, slow, and unnecessary.
Consider the modern developer's workflow: you spin up a Gradio interface for your ML model, a FastAPI backend serving predictions, or a Next.js app consuming those predictions. You want your product manager to see it. Your designer to interact with it. Your client to approve it. Right now, that means deploying to staging, waiting for CI/CD, and dealing with environment mismatches — all for a 5-minute review.
Existing tools were built for a different era. They assume you're a solo dev testing a single webhook. But today, teams are collaborative, AI workflows are complex, and the gap between “it works on my machine” and “let me show you” shouldn't require an ops team.
Requex Tunneling is being built from the ground up with a clear philosophy: it should be as simple as running one command. No account required for basic usage. No config files. No bandwidth throttling. No session limits that cut you off mid-demo. Just a direct, encrypted pipe from the internet to your machine — powered by edge infrastructure that keeps latency low and reliability high.
We believe every developer deserves the ability to make their local work accessible to the world, whether it's a hackathon prototype, a production webhook handler, or a cutting-edge AI model running on a home GPU. That's why we're building this.
Register your interest — we'll notify you when it's ready
Perfect for
ML Engineers & Data Scientists
- Expose a local model API for integration testing
- Let frontend teams hit your GPU-trained model endpoint
- Demo inference results to stakeholders without deployment
- Run Gradio or Streamlit apps and share them instantly
Backend & API Developers
- Test webhook integrations from Stripe, GitHub, Shopify
- Share your local API with mobile developers on the team
- Debug OAuth callback flows without deploying
- Run end-to-end tests against third-party services
Frontend Developers
- Share work-in-progress with clients for instant feedback
- Test responsive design on real devices over the tunnel
- Demo UI changes to the team without staging deploys
- Run integration tests against a locally running backend
Indie Hackers & Side Projects
- Host a personal project temporarily for free
- Give demo access to early beta testers
- Receive webhook events while building on the train
- Prototype and iterate without cloud costs
Register Your Interest
Be the first to know when Requex Tunneling launches. Drop your details and we'll notify you.