Semantic conventions for generative AI systems

Status: Development

[!Warning]

Existing GenAI instrumentations that are using v1.36.0 of this document (or prior):

• SHOULD NOT change the version of the GenAI conventions that they emit by default. Conventions include, but are not limited to, attributes, metric, span and event names, span kind and unit of measure.
• SHOULD introduce an environment variable OTEL_SEMCONV_STABILITY_OPT_IN as a comma-separated list of category-specific values. The list of values includes:
  • gen_ai_latest_experimental - emit the latest experimental version of GenAI conventions (supported by the instrumentation) and do not emit the old one (v1.36.0 or prior).
  • The default behavior is to continue emitting whatever version of the GenAI conventions the instrumentation was emitting (1.36.0 or prior).

This transition plan will be updated to include stable version before the GenAI conventions are marked as stable.

Semantic conventions for Generative AI operations are defined for the following signals:

• Events: Semantic Conventions for Generative AI inputs and outputs - events.
• Metrics: Semantic Conventions for Generative AI operations - metrics.
• Model spans: Semantic Conventions for Generative AI model operations - spans.
• Agent spans: Semantic Conventions for Generative AI agent operations - spans.

Technology specific semantic conventions are defined for the following GenAI system:

• Azure AI Inference: Semantic Conventions for Azure AI Inference.
• OpenAI: Semantic Conventions for OpenAI.
• AWS Bedrock: Semantic Conventions for AWS Bedrock.