Capability domains

The main capability areas covered by AI and the components behind each.

Agentic Friendly

User interface and API

CapabilityMain componentsPurpose
Web experienceAI Web PortalOmniChat for chat, Build, files, Agents, Skills, Connectors, Deep Research and voice; Studio for models, data and compute; and administration
Backend servicesBSQAI API, TemporalTenant-aware application APIs and durable workflows
Developer compatibilityBSQAI APIChat Completions, Responses, Models, Embeddings and Rerank APIs under /v1, and the Anthropic Messages API under /anthropic, for applications, SDKs and coding tools
Application identityBSQAI API, KeycloakAPI keys with a permission scope, and service accounts that give applications their own identity

Model lifecycle and inference

CapabilityMain componentsPurpose
Model registryMLflowTrack experiments, models, versions, and GenAI traces, in the installation's MLflow or an organization's private MLflow
Model deliveryModel InstallerImport, register, and deploy model artifacts
Model servingKServe, vLLM, MLServer, NVIDIA Triton, KAI SchedulerDeploy, schedule, scale, and execute LLMs and predictive models, within each tenant's GPU guarantee and cap
Model accessBSQAI API, OpenFGAPublish models to other organizations and grant their use with allow rules
Usage and budgetsBSQAI API usage records, Analytics, BudgetsAttribute BCU consumption to users, API keys, applications, models and groups, and cap it monthly

Tools, search, and speech

CapabilityMain componentsPurpose
Model routingBSQAI APISend each model request to KServe or to a configured external model server
MCP operationsMCP Gateway, MCP Lifecycle OperatorManage, authorize, and call managed or remote MCP Servers
Web searchBSQAI APIConfigure tenant providers and supply the web_search tool
SpeechSpeachesBatch or real-time transcription with FasterWhisper, and streamed speech generation with Piper and Kokoro voices
Code executionAgent Sandbox, Kata ContainersRun the agent's commands and skill scripts in isolated sandboxes, where the site enables them

RAG and document processing

CapabilityMain componentsPurpose
Document conversionDoclingConvert documents into structured content
Vector searchMilvusStore and query embeddings for tenant-authorized retrieval
Workflow orchestrationTemporalRun durable ingestion, reindex, and document-processing operations

Observability

CapabilityMain componentsPurpose
GenAI observabilityOpenTelemetry, MLflowTrace inference, retrieval, tool, storage, and workflow operations
Telemetry routingGrafana AlloyRoute application and inference signals to Foundation backends

On this page