obsdog

OpenAI integration metrics

The official Datadog integration for monitoring OpenAI. It collects 29 metrics.

Official summary

Optimize OpenAI usage: cost estimates, prompt sampling and performance metrics.

Metric types

  • count (occurrences in the interval): 20
  • gauge (value at a point in time): 9

Metrics

Descriptions are the official definitions, unedited.

MetricTypeUnit
openai.audio_speeches.charactersNumber of characters generated for text-to-speechcount-
openai.audio_speeches.num_model_requestsNumber of text-to-speech model requestscountrequest
openai.audio_transcriptions.num_model_requestsNumber of audio transcription model requestscountrequest
openai.audio_transcriptions.secondsNumber of seconds of audio transcribedcountsecond
openai.code_interpreter_sessions.num_sessionsNumber of code interpreter sessionscountsession
openai.completions.input_audio_tokensNumber of audio input tokens for completionscounttoken
openai.completions.input_cache_write_tokensNumber of input tokens written to the cache for completionscounttoken
openai.completions.input_cached_tokensNumber of cached input tokens for completionscounttoken
openai.completions.input_tokensNumber of input tokens for completionscounttoken
openai.completions.input_uncached_tokensNumber of uncached input tokens for completionscounttoken
openai.completions.num_model_requestsNumber of completion model requestscountrequest
openai.completions.output_audio_tokensNumber of audio output tokens for completionscounttoken
openai.completions.output_tokensNumber of output tokens for completionscounttoken
openai.embeddings.input_tokensNumber of input tokens for embeddingscounttoken
openai.embeddings.num_model_requestsNumber of embedding model requestscountrequest
openai.images.imagesNumber of images generatedcount-
openai.images.num_model_requestsNumber of image generation model requestscountrequest
openai.moderations.input_tokensNumber of input tokens for moderationscounttoken
openai.moderations.num_model_requestsNumber of moderation model requestscountrequest
openai.organization.ratelimit.requests.remainingNumber of requests remaining in the rate limit.gaugerequest
openai.organization.ratelimit.tokens.remainingNumber of tokens remaining in the rate limit.gaugetoken
openai.ratelimit.requestsNumber of requests in the rate limit.gaugerequest
openai.ratelimit.tokensNumber of tokens in the rate limit.gaugetoken
openai.request.durationRequest duration distribution.gaugenanosecond
openai.request.errorNumber of errors.counterror
openai.tokens.completionNumber of tokens used in the completion of a response from OpenAI.gaugetoken
openai.tokens.promptNumber of tokens used in the prompt of a request to OpenAI.gaugetoken
openai.tokens.totalTotal number of tokens used in a request to OpenAI.gaugetoken
openai.vector_stores.usage_bytesNumber of bytes used in vector storesgaugebyte

Official documentation

Official documentation for the OpenAI integration

Metric names and descriptions are the official Datadog integration definitions (Copyright Datadog, Inc., BSD-3-Clause), reproduced unedited. The full license terms are on the About page.

Related products

IntegrationsIntegrations are the bundle of entry points that pour data from external systems such as AWS and MySQL into Datadog. Where you configure them splits into two families, the Agent side and the cloud account side, and which one you choose changes both what you can see and how billing is counted.Foundation and data collectionLLM ObservabilityLLM Observability records a single request to an application with an embedded LLM as a trace, tracks the inputs and outputs, tokens, latency, and errors of each model call, and turns response quality into numbers through evaluations. The only thing billed is the LLM span, which represents a model call.APM and tracingDatadog AgentSoftware that runs resident on the servers and containers you monitor and delivers metrics, logs, and traces to Datadog.Foundation and data collection

Other Cost Management integrations

Anthropic Usage and Costs8 metricsDatabricks16 metricsLangChain6 metrics