Brendan McKeag

Join us as we celebrate our fielding of our 10 billionth serverless request.

Deploy Serverless endpoints directly from GitHub and roll back instantly if needed. Runpod’s improved GitHub integration lets you revert to previous builds without rebuilding Docker images, enabling faster, safer, and more confident deployments.

Runpod has significantly improved the performance and reliability of its automated GitHub integration by fixing a bottleneck in the container image upload pipeline that caused slow or timed-out builds. By rewriting key components of the registry image uploader and optimizing layer transfers, GitHub-triggered container builds now complete faster, more consistently, and with fewer deployment failures.

Runpod has officially achieved SOC 2 Type II certification, validating that its enterprise-grade security controls not only meet strict design standards but also operate effectively over time. This milestone proves Runpod’s ongoing commitment to protecting customer data and maintaining trusted, compliant AI infrastructure for enterprises and developers alike.

Slurm on RunPod Instant Clusters makes it simple to scale distributed AI and scientific computing across multiple GPU nodes. With pre-configured setup, advanced job scheduling, and built-in monitoring, users can efficiently manage training, batch processing, and HPC workloads while testing connectivity, CUDA availability, and multi-node PyTorch performance.

DeepSeek V3.1 introduces a breakthrough hybrid reasoning architecture that dynamically toggles between fast inference and deep chain-of-thought logic using token-controlled templates—enhancing performance, flexibility, and hardware efficiency over its predecessor V3-0324. This update positions V3.1 as a powerful foundation for real-world AI applications, with benchmark gains across math, code, and agent tasks, now fully deployable on RunPod Instant Clusters.

Deploy Wan 2.2 on Runpod to unlock next-gen video generation with Mixture-of-Experts architecture, TI2V-5B support, and 83% more training data—run text-to-video and image-to-video models at scale using A100–H200 GPUs and customizable ComfyUI workflows.