Skip to main content
Anaconda Platform 7.0.0 is available through a limited early access program. Contact your Anaconda Technical Account Manager (TAM) if you’re interested in adopting the latest version.
If you are using Anaconda’s cloud-based platform, select Cloud from the dropdown beside the Anaconda logo at the top of the page.
Anaconda Platform (Self-hosted) serves as the unifying foundation for organizations to centrally manage and govern the open-source software, AI models, and computational resources their teams rely on. By integrating package management, model hosting, and policy enforcement within a single, secure environment, the platform empowers administrators to enforce compliance, streamline workflows, and ensure that every resource meets organizational standards for trust and security. This comprehensive approach enables teams to innovate confidently while maintaining governance at scale.

​
Key Features

​
Anaconda Core

Anaconda Core is the backbone of Anaconda Platform (Self-hosted), providing structured governance and centralized management of users, groups, channels, mirrors, and packages across your organization.
Identity and Access Management
Administrators can create and organize users into groups, assigning role-based permissions that provide access to most features of the platform. This granular access control ensures that individuals and teams have the precise permissions they need, maintaining both security and clarity across the organization.
Package Governance
Anaconda Core provides fine-grained control over which packages enter your organization’s channels from mirrored repositories. Administrators define filters and policies that determine which packages are allowed, ensuring teams only have access to software that meets organizational security and compliance standards.

​
AI Catalyst

AI Catalyst extends Anaconda Platform (Self-hosted) by providing access to and governance of AI models through Anaconda’s curated model repository.
Model Governance
Set organization-wide restrictions to block unapproved models and assign group permissions to make specific models available for approved users. These governance rules ensure that teams only use models that meet your organization’s standards for security, compliance, and performance.
Catalog Synchronization
Keep your local model catalog up to date with the latest models being added to Anaconda’s curated repository. When governance rules or permissions change, synchronization downloads newly approved models and removes any that are no longer allowed. Synchronization can be manually triggered or set to run on a scheduled basis.
Managed Model Servers
Host approved models on platform-managed servers with reliable, scalable endpoints for inference and embeddings. This allows teams to deploy and use governed models securely in production environments.

​
JupyterHub integration

Anaconda Platform (Self-hosted) integrates with JupyterHub to provide a centrally managed, secure notebook workspace for your organization. User accounts and groups are automatically synchronized from the platform, allowing members to sign in using their existing Anaconda Platform (Self-hosted) credentials. During installation, administrators can configure per-user compute limits and define which channels from Anaconda Platform (Self-hosted) are available within JupyterLab. Notebooks created in JupyterHub automatically reference these approved channels when building environments, ensuring every workspace is consistent, reproducible, and compliant with organizational policies.

​
Monitoring and reporting

Anaconda Platform (Self-hosted) reports on both product usage and system health. Administrators can review user activity and audit logs to see who performed which actions and when, and they can generate support bundles for troubleshooting purposes. System metrics are collected by Prometheus and Alertmanager and visualized in Perses dashboards, covering resource consumption, pod readiness, and alerts. This gives teams a clear picture of how the platform is being used and how it is performing.