Monitor Your OpenAI Usage With Datadog

Secure Your Cloud Infrastructure

Assess and enhance your security posture with continuous configuration checks across your cloud accounts, hosts, and containers
Track and fulfill audit requirements with out-of-the-box support for over 15 compliance frameworks including PCI DSS, SOC 2, HIPAA, GDPR, and CIS
Automatically detect and prioritize cloud vulnerabilities based on potential business impact, deep observability context, and Datadog’s curated security research
Visualize your cloud resources and their associated security risks at a glance with a comprehensive resource inventory

Easily break down API requests by OpenAI model, service, organization ID, and API key with out-of-the-box dashboards
Identify where OpenAI’s models are used in your organization and to what extent with unique organization IDs
Enable tracking by API keys to break down usage by specific users, uncover spikes in usage and associated costs, and trace unauthorized API access back to specific keys and users
Avoid rate-limit errors or excessive latencies with recommended monitors that proactively measure the rate of your requests and token usage

Discover primary cost drivers for your OpenAI usage; track your total token consumption and break it down by prompt and completion tokens
Leverage out-of-the-box dashboards to track the average total number of tokens per request & and the average number of prompt and completion tokens per request
Deploy and start monitoring without any need for professional services or extensive training

Seamlessly track the ratio of response times to the volume of prompt tokens and distinguish between anomalous latencies from prolonged response times caused by spikes in requests
Utilize traces to uncover performance issues on individual requests, such as the specific models that were queried and the precise content of prompts and completions
Automatically save error and high latency traces for 15 days

Easily query observability data in natural language when and where you need it on mobile, web, or Slack
Accelerate investigations and improve collaboration with automatically generated incident summaries, and post mortems
Democratize observability data and accelerate onboarding with your always-on generative AI Copilot
Resolve bugs faster in production and fix code quality issues with automatically generated code fix suggestions and unit tests