There is plenty of computational resources (both GPU, CPU and disk space).
Saturn Cloud offers lightning-fast CPUs and supports GPUs for machine learning, allowing users to edit environments before starting resources. It scales according to user needs and provides a centralized space for data, code, and results. However, limitations exist in library installation, documentation detail, and hardware options. Public Clouds integration and sandbox environments are desired, and pricing revision, similar to AWS spot instances, is suggested. Expanding to include TPU coprocessors or ARM64 CPUs is requested.