Monitor and track performance of your OpenResponses deployment
TelemetryService
.
This service provides methods to:
Metric Name | Description |
---|---|
execute_tool | Measures internal tool execution performance |
gen_ai_client_operation_duration | Tracks duration of model API calls |
gen_ai_client_token_usage | Counts input and output token usage |
/v1/responses
or /chat/completions
{gen_ai.operation.name} {gen_ai.request.model}
execute_tool