# What is opentelemetry genai semantic conventions?

> OpenTelemetry GenAI semantic conventions are the standard names for telemetry from generative AI systems: span names, gen_ai.* attributes, metrics, and events for LLM inference, embeddings, agents, tool calls, and MCP, so any OpenTelemetry backend can read LLM and agent traces the same way.

- HTML: https://openlit.io/glossary/opentelemetry-genai-semantic-conventions
- Markdown: https://openlit.io/glossary/opentelemetry-genai-semantic-conventions.md

An LLM inference span is named "{gen_ai.operation.name} {gen_ai.request.model}", for example "chat gpt-4". The span carries gen_ai.provider.name (required, such as openai or aws.bedrock), gen_ai.request.model, and token counts in gen_ai.usage.input_tokens and gen_ai.usage.output_tokens. Agent spans use operations such as create_agent and invoke_agent, and tool calls use execute_tool. The conventions also define spans and metrics for MCP clients and servers.

The conventions are still marked Development, so attribute and metric names can change between releases. In 2026 they moved from the main semantic-conventions repository to a dedicated repository, open-telemetry/semantic-conventions-genai, which uses OpenTelemetry Weaver to generate the docs. Instrumentation should pin a semantic conventions version and expect migrations.

OpenLIT SDKs emit spans and metrics that follow the GenAI semantic conventions for LLM providers, vector databases, and agent frameworks, with a single openlit.init() call. Because the output is standard OTLP, the same traces work in OpenLIT, Grafana, Datadog, or any OpenTelemetry backend without re-instrumenting.

## FAQ

### What are OpenTelemetry GenAI semantic conventions?

They are the OpenTelemetry standard for naming generative AI telemetry: gen_ai.* span attributes, metrics, and events for LLM calls, embeddings, agents, tool calls, and MCP. They let any OTLP backend understand LLM and agent traces.

### Are the GenAI semantic conventions stable?

No. They are marked Development, so names can change. They are maintained in the open-telemetry/semantic-conventions-genai repository.

### Which tools emit OpenTelemetry GenAI semantic conventions?

OpenLIT, OpenLLMetry, and the OpenTelemetry Python contrib instrumentations emit gen_ai.* spans, and backends such as OpenLIT, Grafana, and Datadog can ingest them. OpenLIT is open source under Apache 2.0 and instruments LLMs, vector databases, and agent frameworks with one line of code.

## Related

- [LLM observability](https://openlit.io/glossary/llm-observability.md)
- [Agent observability](https://openlit.io/glossary/agent-observability.md)
- [Agent harness](https://openlit.io/glossary/agent-harness.md)
- Pillar: https://openlit.io/agent-harness-engineering.md
