GPUs are a critical tool for tasks like machine learning, data processing, and high-performance computing, but integrating them into CI/CD pipelines presents unique challenges. These challenges include the cost and accessibility of GPUs, which can be expensive and difficult to share across containers, as well as limited use cases that often lead to underutilized GPU resources. To simplify GPU integration, Dagger is making it easier to use remote runners like Fly.io and Lambda Labs to run GPU-enabled pipelines on-demand, saving time, cost, and headaches. With the latest updates to its documentation, users can offload GPU-specific tasks to these remote runners while keeping the rest of their pipeline local, deploying resources only when needed, persisting caching across runs, and managing infrastructure agnosticity. To see GPU integration in action, Dagger provides a demo that walks through a practical example using Fly.io and Lambda Labs, allowing users to integrate GPU steps seamlessly without breaking their existing workflows.