Join us for our first Flash hackathon! June 30, SF. Remote submissions welcome. luma.com/runpod-flash-jun-30

"Which inference engine should we use, vLLM or TensorRT-LLM?" We see this question a lot. If you're mid-decision, the full breakdown is here, including where each one tends to surprise people: runpod.io/articles/compariso…

Deploy When Available is now GA. Queue for any GPU spec, even if it's fully rented out, and we'll deploy your pod the moment it frees up. Here's how to get started → runpod.io/blog/deploy-when-a…

Flash Hack Day is in the books. 11 teams hacked and demoed live, with real production endpoints, shipped in a day. Congrats to Phong Cao (FlashML), Nelson Lai (FlashDock), and Sulaxmi Raskar (LifeLine). First one, not the last. You can find all projects here: hackersquad.io/events/runpod…

most inference optimization: here are 20 popular models we made faster runpod overdrive: what model are you actually running runpod.io/overdrive?utm_sour…

We'll be at the AI Engineer World's Fair this week, June 29 – July 2 in San Francisco. If you're attending, come find us! We always love to chat about what you're building. See you there.

Two days until Flash Hack Day. We've got credits and cash prizes for the top projects, plus a gift for every single person who builds with us. The event is remote-friendly. So if you can't join us in San Francisco, just join us online! Register through the link below: luma.com/runpod-flash-jun-30

Make Tuesday the best day of your week. Spend the day building with us on June 30th in SF. Registration link is below: luma.com/runpod-flash-jun-30

Today, we’re excited to announce our $50M Series B, led by @GreenfieldVC, with participation from @lightspeedvp and @notablecap. 🚀 At Patronus AI, we develop simulations and evals to train and improve AI. The first phase of AI was built on static benchmarks, but that era is over. As agents are used to solve longer and longer tasks, they need to practice in dynamic, living worlds to get better. Simulations are the critical infrastructure powering this next phase. As a company, we’re behind the most influential research and products in AI evaluation, like FinanceBench, Lynx, and Percival. And things have moved at the speed of light since.⚡ We partner with the world's leading frontier AI labs and enterprises, and our revenue has grown more than 15x over the past year. Additionally, today, we’re introducing a preview of the first Digital World Model for AI agent training and simulation: Patronus-DWM. Digital World Models are language diffusion world models that predict realistic environment behaviors and steer agent actions across digital workflows. Just as physical world models predict how objects move through space, we’re developing the equivalent for the digital world: predicting how agents act in digital workflows, then using that to scale the creation of high-quality training data for LLMs. Digital World Models help us push the frontier of ultra long horizon workflows, and unlock a new class of self-improving RL environments. This is our scalable approach to simulating all of the world’s intelligence. The round was also joined by @datadoghq, @SamsungVentures, @gokulr, @factorialcap, and a large cohort of amazing AI leaders across @AnthropicAI, @OpenAI, @GoogleDeepMind, @nvidia, @Recursive_SI, and more.✨ It has been the ride of a lifetime. But we’re just getting started. The best is yet to come. "Do not go gentle into that good night, Rage, rage against the dying of the light" - Dylan Thomas (1954)

We just closed a $100M Series A led by @SummitPartners. 🎉 1M developers. $22M raised to get here. Almost entirely word of mouth. Pods. Serverless. Clusters. The full AI lifecycle on one platform. We're moving faster on all of it. runpod.io/blog/one-million-d…

Your GPU has a ceiling. Most workloads don't hit it. That's the gap that @Recursive just posted about. Worth a read! recursive.com/articles/first…

We had a great time at the @databricks Data AI Summit this week. Always great to see so many builders who are shipping real AI workloads. Thanks to everyone who stopped by the booth, and we'll see you at the next one!

What happens before the first request hits your serverless endpoint? Join us on June 25 when Greg Wester shows you a live deployment of FlashBoot & batch inference. Register here: resources.runpod.io/whats-ne…

Join us June 25 for a live demo of what's new in Runpod Serverless — hot starts, batch inference, and a cold-to-running deployment on stage. Register here: resources.runpod.io/whats-ne…

More than half of all internet traffic is now AI agents. Behind every one of those requests is an inference call that needs to cold-start fast, scale under load, and cost nothing at idle. That's the infrastructure problem we're solving at Runpod.

Come build with us during our Flash Hack Day in SF on June 30. Expect a full-day hackathon where you ship real projects. No Dockerfile, no container registry, no infra config. Yes, just Python! Register here: luma.com/runpod-flash-jun-30

Agents have overtaken humans in internet traffic. And behind every agent is an inference call that needs to be fast, cheap, and scalable. That's what we built Runpod for. Deploy in under 30 seconds. Scale from zero. Pay nothing at idle.

The most costly inference failures usually aren't in your model. They're at the boundary between two systems that were never designed to talk to each other. Remove the seams and the failures go with them. That's how Scatter Lab runs 1,000 RPS in production on Runpod. runpod.io/case-studies/how-s…

Getting GPUs right now is already hard. Once you finally have them, you're staring down node networking, cross-machine scaling, and failure recovery. The work nobody warned you about. Compute and software should be designed together. That's what we're building.