Model Training

The MODEL TRAINING section in the Admin UI sidebar groups AIOps pages in OptScale AI into two documentation areas:

Documentation	Sidebar pages	Read more
Experiment Tracking	Tasks, Models, Datasets, Artifacts	Experiment Tracking
Environments & Operations	Shared Environments, Cloud Connections, Power Schedules, Integrations	Environments & Operations

Use Experiment Tracking to run and compare training workloads, register models, version datasets, and store run outputs. Use Environments & Operations to book shared runtimes, link cloud accounts, schedule instance power, and connect calendars or CI/CD.

For AI Gateway, policies, MCP servers, and FinOps pages above this section, see Core Services. For platform architecture, see Architecture Overview.

Menu section overview

Table 1: OptScale AI Model Training — sidebar menu items

Menu item	Purpose
Tasks	Create, run, and monitor training or pipeline executions.
Models	Register and manage model versions, metadata, and lifecycle stages.
Datasets	Organize and version datasets used by training and evaluation runs.
Artifacts	Store run outputs such as checkpoints, logs, and generated files.
Shared Environments	Manage shared runtime environments used by teams for training and testing.
Cloud Connections	Configure cloud account integrations used by environments and compute resources.
Power Schedules	Define automated start and stop schedules for infrastructure resources.
Integrations	Connect external systems and CI/CD services used in ML workflows.

Related documentation

Topic	Where to read more
Experiment Tracking (tasks, models, datasets, artifacts)	Experiment Tracking
Environments, cloud, schedules, CI	Environments & Operations
AI Gateway, policies, MCP, FinOps	Core Services
Platform architecture and request flow	Architecture Overview
Initial platform setup	First Steps
Chat workspace	Interface Overview