ClearML

ClearML is an AI infrastructure platform for machine learning and generative AI teams. It is organized in three layers. The Infrastructure Control Plane connects and manages GPU clusters across on premises, cloud and hybrid environments, with quota management, priority based job scheduling, fractional GPUs, multi tenant isolation, role based access control and usage based billing. The AI Development Center is a workbench for building, training and testing models, covering experiment management, dataset versioning, a model repository, pipelines, agent orchestration and CI/CD automation. The GenAI App Engine deploys large language models onto a team's own clusters and handles networking, authentication and security. The platform is aimed at AI builders and the IT teams that supply their compute, from individual researchers to enterprises with large GPU estates. It can be used on the vendor's hosted servers, self hosted, or run as a managed service in a VPC, on premises including air gapped sites, or in a hybrid setup. On hosted servers the Community plan is free for teams up to 3, with 100GB of artifact storage, 1GB of metric events and 1M API calls per month. Pro is $15 per user per month plus usage for teams up to 10, adding cloud auto scaling on AWS, GCP and Azure, hyperparameter optimization, pipeline triggers and dashboards, with listed overage rates such as $0.1 per GB of artifact storage and $1 per 100K API calls. Scale, for organizations with 8 to 48 GPUs, and Enterprise are quoted on request.