ngrok
@ngrok.com
One gateway for all your traffic. Service status available at http://status.ngrok.com.
You should be able to send inference: → to any public provider → privately to models you self-host → to any number of fallbacks → with full access control and observability built in ngrok.ai is live on Product Hunt today ↓ www.producthunt.com/products/ngr...
One gateway, every model — ngrok.ai
Route, secure, and manage traffic to any LLM—cloud or local—with one unified platform. Monitor usage, optimize costs, and keep your AI products online.
ngrok.ai
"You need four of this pod... how hard can that be?" — @samwho.dev Turns out it's several thousand lines of code hard. Part 4 of our video series on his Kubernetes-in-the-browser project tiptoes through that and other footguns ↓ youtu.be/y4XAOcCT9dY
Webernetes: How we got Kubernetes running in the browser
YouTube video by ngrok
youtu.be
💡 Tuesday tech tip: At our free & virtual #P99CONF, @ngrok.com's Sam Rose will dive into the mathematical process applied to models to make them smaller and introduce our audience to tools that gauge how much a quantized model differs from its original. www.p99conf.io?latest_sfdc_... #ScyllaDB
Validating JWTs, restricting IPs, and running a WAF on your ngrok endpoints just got 10x cheaper. Traffic Policy is now $0.10 per 100k TPUs, down from $1. OAuth didn't get cheaper. It was already free. ↓ ngrok.com/blog/pricing...
Traffic Policy is now 10x cheaper, and idle custom domains are free | ngrok blog
Traffic Policy pricing drops by 10x, custom domains move to active-only billing, and Endpoint Pooling now bills per active endpoint. Here's what changed and why.
ngrok.com
The next interactive essay from @samwho.dev is about k8s probes, but even starting it took… 90k lines of TypeScript. Here's webernetes episode 3: the first demo of a cluster in a browser tab. Deployment controllers due in 3 days, when Sam's summer holiday starts ↓ www.youtube.com/watch?v=gxo3...
The demo that proves Kubernetes can run in the browser
YouTube video by ngrok
youtube.com
You can now call models running on your own hardware through a hosted AI gateway. Privately, with one command: `ngrok http 8000 --url https://vllm.internal` Plus all the public models you know and love. ↓ ngrok.ai
Route, Secure & Manage Any LLM with ngrok AI Gateway
Route, secure, and manage traffic to any LLM—cloud or local—with one unified platform. Monitor usage, optimize costs, and keep your AI products online.
ngrok.ai
I promise this is my last post this week sharing work stuff but we’ve condensed a bunch of new stuff @ngrok.com can do into a 6 and a half minute video! This is for the folks that know how to run `ngrok http 8080` but haven’t ventured further into the product than that. 🫶 youtu.be/trXqyNXJa1k?...
This is ngrok in 2026
YouTube video by ngrok
youtu.be
Yes, ngrok gives localhost a public URL. It's also load-bearing production infrastructure. Reach into customer networks. Put every bit of ingress behind one front door. Run one gateway for devices, APIs, and LLMs. Here's what ngrok really is today ↓ www.youtube.com/watch?v=trXq...
This is ngrok in 2026
YouTube video by ngrok
youtube.com
Inviting controversy with this take, which is outside of my comfort zone, but I do stand by it. You can generate good quality code with LLMs, it “just” requires a tonne of human oversight and automated guard rails. Got a week off then I’ll be building my first proper post with webernetes. Stoked.
ngrok@ngrok.com · last mo.Our own @samwho.dev ported Kubernetes to the browser. Like a real flippin' cluster with lifecycles, DNS, and a simulated network. It's ~100k lines of TypeScript, almost all written by LLMs, but with every line reviewed by hand to keep it slop-free ↓ ngrok.com/blog/i-porte...
Our own @samwho.dev ported Kubernetes to the browser. Like a real flippin' cluster with lifecycles, DNS, and a simulated network. It's ~100k lines of TypeScript, almost all written by LLMs, but with every line reviewed by hand to keep it slop-free ↓ ngrok.com/blog/i-porte...
I ported Kubernetes to the browser | ngrok blog
Almost 100,000 lines of LLM-generated code in 2 months, and none of it is slop.
ngrok.com
Yup, @samwho.dev is a wild one for this. But we want to teach the invisible layers of modern software in ways grounded in reality. Ways that'll last for a long, long time. Sometimes that means 100k lines of TypeScript just to earn the right to start telling the story at all.
Sam Rose@samwho.dev · 2mo agoI've made a Kubernetes that runs in the browser, and today we're open sourcing it. The hope here is to make super interactive and visual web content about Kubernetes much easier to write, and we have a whole bunch planned this year. Repo: github.com/ngrok/webern... Demo: webernetes-demo.ngrok.app
Our very own @samwho.dev decided the only honest way to explain how Kubernetes probes work, using his signature interactive essay style, was to build himself a cluster. In the browser. 👇
We gave ngrok domains a `resolves_to` property and, what do you know, made the *routing* part of data residency a one-click. Pin your traffic to specific PoPs so European data stays in Europe, then rearrange the whole thing tomorrow if you want.
Expose local apps and APIs to the internet with your coding agent and our new `expose-localhost` skill. Then ask it to add in OAuth, OWASP protection, rate limits, and more if you're feeling plucky. $ npx skills add ngrok/agent-skills github.com/ngrok/agent-...
GitHub - ngrok/agent-skills: Official repository for ngrok agent skills.
Official repository for ngrok agent skills. Contribute to ngrok/agent-skills development by creating an account on GitHub.
github.com
Quantization can make an LLM 4x smaller and 2x faster, with barely any quality loss. But what *is* it? @samwho.dev crafted a beautiful interactive essay explaining it from first principles, aimed at coders, not mathematicians. ngrok.com/blog/quantiz...
Quantization from the ground up | ngrok blog
A complete guide to what quantization is, how it works, and how it's used to compress large language models
ngrok.com
I spent 2 months learning about quantization and am extremely proud of the post I've written about it. I think these are some of the nicest visuals I've ever made, and I love how this compression technique invented in 1898 is being used on the bleeding edge in 2026. ngrok.com/blog/quantiz...
Quantization from the ground up | ngrok blog
A complete guide to what quantization is, how it works, and how it's used to compress large language models
ngrok.com
Route between OpenAI and Anthropic with one API key without an account for either. ngrok's AI Gateway now has its own API keys and prepaid credits. One key covers both, and you can just set your model to "ngrok/auto" if you don't even want to pick. ngrok.com/blog/ai-gate...
AI Gateway: use Anthropic and OpenAI with one API key | ngrok blog
Create an AI Gateway API Key, add credits, and start making requests to OpenAI and Anthropic—ngrok handles the rest.
ngrok.com
ngrok's AI gateway now supports the Anthropic SDK natively. Change your base_url, keep your prompt caching, extended thinking, all of it. The diff is one line & there's no step two. ngrok.com/blog/native-...
What do LLMs see? I wrote a lil' tool that extracts the attention matrices out of open models and creates this typing visual, with each token's opacity changing according to its average attention score as the prompt progresses. Dimmer words are considered less important to the model.
This week on Overcommitted, we got to sit down with Bluesky's favorite tech blogger @samwho.dev and it did not disappoint! Sam makes some of the coolest tech content on the internet, and if you haven't heard from him yet, you should! Full episode out now: overcommitted.dev/interactive-...
We're hiring a full-time video creator. Looking for someone who always aims for quality above cadence, is eager to deep-dive on AI and networking, wants to be part of a small and taste-obsessed team. Details: job-boards.greenhouse.io/ngrokinc/job...
Sr. Developer Educator, Video
United States
job-boards.greenhouse.io
Did you know that SWE-bench only tests how good models are at Python? Or the big drama around the FrontierMath benchmark? We didn't either! But now you will, because @samwho.dev put together an extensive explanation of 14 popular AI benchmarks. ngrok.com/blog/ai-benc...
What those AI benchmark numbers mean | ngrok blog
An explanation of 14 benchmarks you're likely to see when new models are released.
ngrok.com
Three ships coming soon we're excited about: ◆ region pinning for data residency ◆ similarly, dedicated IPs instead of our default range of multi-tenant IPs ◆ faster & more reliable certificate provisioning for ngrok/custom/wildcard domains ↓
ngrok is one gateway for all your traffic. So, time to introduce the new ngrok.com For years we've helped you secure, transform, and route to services running anywhere. localhost to prod, APIs to AI models, devices in the field to databases in customer networks.
ngrok - All your traffic. One gateway. | API Gateway, Secure Tunnels, Traffic Management
ngrok is an all-in-one cloud networking platform that secures, transforms, and routes your traffic to services running anywhere.
ngrok.com
We heard you like paging through dozens or hundreds of objects in API responses to find the one you want. Just kidding, nobody's ever said that, so thank goodness that server-side filtering of the ngrok API is now GA. ngrok.com/blog/api-fil...
An enjoyable look at how LLMs work under the hood from @samwho.dev. The whole chain of tokenizers (text chunks), embeddings (vectorized text chunks), transformers (the "T" in ChatGPT), caching (reused chunks), and all the math that goes into returning a response. ngrok.com/blog/prompt-...
Prompt caching: 10x cheaper LLM tokens, but how? | ngrok blog
A far more detailed explanation of prompt caching than anyone asked for.
ngrok.com
Learned more about LLM's under the hood from this post than my stupid $30 online course ngrok.com/blog/prompt-... @samwho.dev doing @samwho.dev things 🙌
Prompt caching: 10x cheaper LLM tokens, but how? | ngrok blog
A far more detailed explanation of prompt caching than anyone asked for.
ngrok.com
New post! ✨ Prompt caching ✨ My first big project post for @ngrok.com. 5 weeks, 12217 lines of code, 195 commits. I poured a lot into this one, and learned a lot in the process. I really hope you enjoy it ❤️
ngrok@ngrok.com · 8mo agoYesterday we launched ngrok.ai into early access. Today we're bringing you a deep dive into LLM internals with beautiful visuals crafted by our very own @samwho.dev. Discover exactly what gets cached to offer you 10x cheaper input tokens. ngrok.com/blog/prompt-...
Yesterday we launched ngrok.ai into early access. Today we're bringing you a deep dive into LLM internals with beautiful visuals crafted by our very own @samwho.dev. Discover exactly what gets cached to offer you 10x cheaper input tokens. ngrok.com/blog/prompt-...
