Hello HN<p>I’ve spent too many hours reading GPU providers docs, renting machines, setting them up, fixing dependencies just to get a single AI model online. So a few friends and I built RunGen.AI, a platform where you paste a link (Hugging Face or CivitAI), click create and you’ve got a live API endpoint in minutes.
We’re still in beta, but here’s what’s working so far:<p>* Bring your own model: No vendor lock-in.<p>* Pay per GPU minute<p>* Minimal setup: No infra to manage, no complex configs.<p>Over the next few months, we’re adding private model deployments, autoscaling, GPU location preference (EU/US), and faster startup times. Would love to hear any feedback,what works, what doesn’t, and what you’d like to see next.<p>Check it out at platform.rungen.ai, and feel free to reach us at hi [at] rungen [dot] ai.