Docs

Azure OpenAI

Ingest Azure OpenAI traces into AITracer via the SDK or API.

Agent Lab: Run local Ollama agents in the dashboard — no SDK required (Quick Start). This guide is for ingesting traces from this provider via the SDK or API.

Centralize Azure OpenAI runs in AITracer with trace.record or POST /api/traces after each deployment call.

Captured fields match other OpenAI-compatible ingest:

deployment name / model
prompt and response
token usage and latency
action name and environment

Azure execution workflow

Rendering diagram...

Common patterns

Enterprise copilots

Tag traces with actionName per product (e.g. hr-copilot, support-assistant) so teams can search and export examples for fine-tuning.

Cost monitoring

Include token metrics on every ingest to populate Cost Intelligence — useful when many deployments share one subscription.

Regional deployments

Set environment or metadata.region so you can filter traces by geography without separate AITracer instances.

Optional advanced

Entra ID, RBAC, and audit programs may additionally use Governance Engine, Verification Layer, and Audit Vault when configured on ingest.

Terminal agent client Slack