# Attrs

> **Class** in `langchain_azure_ai`

📖 [View in docs](https://reference.langchain.com/python/langchain-azure-ai/callbacks/tracers/inference_tracing/Attrs)

Semantic convention attribute names used throughout the tracer.

Attributes sourced from the OpenTelemetry GenAI semantic conventions SDK
(``opentelemetry.semconv._incubating.attributes.gen_ai_attributes``).
Custom attributes not yet in the SDK are defined inline.

## Signature

```python
Attrs()
```

## Properties

- `PROVIDER_NAME`
- `OPERATION_NAME`
- `REQUEST_MODEL`
- `REQUEST_MAX_TOKENS`
- `REQUEST_TEMPERATURE`
- `REQUEST_TOP_P`
- `REQUEST_TOP_K`
- `REQUEST_STOP`
- `REQUEST_FREQ_PENALTY`
- `REQUEST_PRES_PENALTY`
- `REQUEST_CHOICE_COUNT`
- `REQUEST_SEED`
- `REQUEST_ENCODING_FORMATS`
- `RESPONSE_ID`
- `RESPONSE_MODEL`
- `RESPONSE_FINISH_REASONS`
- `USAGE_INPUT_TOKENS`
- `USAGE_OUTPUT_TOKENS`
- `TOKEN_TYPE`
- `INPUT_MESSAGES`
- `OUTPUT_MESSAGES`
- `SYSTEM_INSTRUCTIONS`
- `OUTPUT_TYPE`
- `TOOL_NAME`
- `TOOL_TYPE`
- `TOOL_DESCRIPTION`
- `TOOL_DEFINITIONS`
- `TOOL_CALL_ID`
- `TOOL_CALL_ARGUMENTS`
- `TOOL_CALL_RESULT`
- `DATA_SOURCE_ID`
- `AGENT_ID`
- `AGENT_NAME`
- `AGENT_DESCRIPTION`
- `CONVERSATION_ID`
- `REQUEST_MAX_INPUT_TOKENS`
- `REQUEST_MAX_OUTPUT_TOKENS`
- `USAGE_TOTAL_TOKENS`
- `SERVER_ADDRESS`
- `SERVER_PORT`
- `ERROR_TYPE`
- `RETRIEVER_RESULTS`
- `RETRIEVER_QUERY`
- `EVALUATION_NAME`
- `EVALUATION_SCORE_VALUE`
- `EVALUATION_SCORE_LABEL`
- `EVALUATION_EXPLANATION`
- `EVALUATION_RESULT_EVENT`
- `AGENT_STATE`
- `AZURE_MONITOR_CUSTOM_EVENT_NAME`
- `OPENAI_REQUEST_SERVICE_TIER`
- `OPENAI_RESPONSE_SERVICE_TIER`
- `OPENAI_RESPONSE_SYSTEM_FINGERPRINT`

---

[View source on GitHub](https://github.com/langchain-ai/langchain-azure/blob/7a6a664d617264df8dcb12fcfb1bacc09f21d504/libs/azure-ai/langchain_azure_ai/callbacks/tracers/inference_tracing.py#L161)