Introduction#

What is OptScale AI#

OptScale AI is a unified AI operations platform that combines AI Gateway, FinOps, profiling, and observability. It helps teams manage access to AI models, monitor token usage and cost, apply governance controls, and observe AI workloads from one place.

Administrators use the Admin Console to configure providers, access controls, policies, guardrails, and MCP servers. Users interact with approved models through Chat, APIs, and external tools such as OpenCode and Cursor.

Chat, API clients, and compatible external tools use the same governed access layer. Requests are processed according to the organization's provider access, policies, guardrails, and other governance settings.

Dashboards, traces, and usage views help teams monitor AI requests, cost, performance, and platform activity.

For key platform terms and capabilities, see Core Concepts. For platform structure and request processing, see Architecture Overview.

Main use cases#

Centralize AI access. Provide teams with a single, governed way to access configured AI providers instead of relying on separate provider credentials and disconnected integrations.

Control cost and usage. Monitor token usage and cost across providers and teams. Use dashboards and usage views to understand consumption, identify inefficiencies, and track spending.

Enforce governance and compliance. Apply policies and guardrails to control how AI requests are evaluated and handled according to organization requirements.

Operate multi-provider AI workloads. Use multiple AI providers and models from a common platform. Configure routing, fallbacks, and traffic distribution for workloads that require greater resilience or flexibility.

Extend AI workflows with approved tools. Connect MCP servers so Chat and API workflows can use approved external tools and data sources.

Monitor and troubleshoot AI requests. Use traces, dashboards, and observability views to investigate failures, latency, provider behavior, and request processing.

Work with AI through Chat and external clients. Use Chat for interactive AI work, or connect compatible external clients such as OpenCode and Cursor through the OptScale AI Gateway.

You can also migrate existing OpenAI-compatible applications to OptScale AI by updating their connection settings. See Switch to OptScale AI Gateway.

To bring existing conversation history into Chat, see Migrate Chat History.

Supported AI providers#

OptScale AI connects to commercial AI APIs, cloud AI platforms, and self-hosted inference endpoints.

Supported integrations include providers such as OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI, Mistral AI, Ollama, vLLM, and others.

Administrators configure provider connections under Config → Providers. The providers and models available to users depend on the connections enabled for the organization and the user's provider access.

For the full list of supported API types and configuration options, see Config → Providers.

Next steps#

Choose the next guide based on what you want to do:

• First Steps — Configure an organization and provider, then send your first AI request.
• Core Concepts — Learn the main OptScale AI components and terminology.
• Architecture Overview — Understand platform structure and request processing.
• Admin Console Overview — Explore administrative areas and configuration.
• Interface Overview — Learn the Chat interface.
• External Tools — Connect compatible external AI clients.