Skip to main content

Welcome

Orchestrate and customize AI workloads on any infrastructure, at the scale you need


Clarifai is a hybrid cloud AI orchestration platform for building, deploying, and managing AI at scale. The platform supports AI models developed with Clarifai's own set of tools or third-party frameworks, integrating community, private, and self-hosted models.

With Clarifai, teams can deploy AI workloads on our Shared Compute, Clarifai-managed dedicated cloud compute, or within their own VPCs, on-premise, hybrid, or edge envirnoments from a unified control plane.

Clarifai is designed for three core journeys:

  • Compute – Orchestrate AI models with cost-efficient deployment, scaling, and optimization across environments.
  • Control & Governance – Manage AI resources, monitor performance, and track costs through a unified control plane.
  • Create & Manage – Customize AI workloads with a set of tools to compliment your own, enabling you to label data, train and evaluate models, and deploy them.

Compute

  • Deploy AI Anywhere — Run AI on any hardware, cloud, or on-premise infrastructure, including private VPCs or Clarifai’s managed compute.

  • Optimize Costs — Leverage GPU fractioning, auto-scaling including scale-to-zero, and batched or streaming inference to more efficiently utilize compute resources.

  • Enterprise-Grade Security — Manage access and permissions with role-based access controls, organizations and teams, and audit logging.

  • Scale to Production — Deploy models with push-button builds and automated scaling.

Unified Control and Governance

  • Monitor Performance — Track model activity and system performance in real time.

  • Manage Cost — Monitor resource usage and expenses across deployments.

  • Audit Activity — Review time-stamped logs for updates and system changes.

  • Controll Acess — Manage permissions and control who can create, update, or delete AI resources.

Create and Manage

  • Automated Data Labeling — Auto-label images, videos, and text faster and more accurately.

  • Vector Search — Search unstructured data with high precision.

  • Resource Versioning — Track changes to datasets, models, and workflows by versioning your AI assets.

  • Dataset Management — Ingest, exlore, and manage your datasets for model training, model evaluation, and data labeling.

  • Model Training & Evaluation — Train AI models using one-click training templates and make custom AI accessible across teams.

  • AI Workflows — Combine multiple models and operators with a drag-and-drop interface.

  • UI Modules — Extend functionality with reusable web apps leveraging our APIs.