Monitor AI Gateway ​

The AI Gateway dashboard provides visibility into the traffic flowing through your gateway. Use it to monitor health, attribute spend, identify top consumers, and verify that budgets behave as expected.

The dashboard is scoped to the current workspace and includes a time range filter.

AI Gateway dashboard

Overview metrics ​

The dashboard header surfaces summary tiles for high-level metrics across the selected time range, including request volume, token consumption, time to first token (TTFT), and success and failure rates.

Two pie charts break down activity by source:

• LLM Provider Usage Count: Request share per provider, such as OpenAI, Anthropic, Azure OpenAI, and AWS Bedrock.
• LLM Model Usage Count: Request share per model.

Token usage over time ​

This section tracks token consumption across time:

• Token Usage: Total tokens consumed per month.
• Cumulative Token Usage by Provider: Total tokens per month, organized by provider.

Highest token consumption ​

The Top Consumers by Token Usage chart ranks API keys by total tokens consumed. Use it to identify which API keys, teams, or applications drive the most spend.

Cost ​

The Budget vs Actual Spend chart compares actual spend against the configured budget over time. Use this chart to drill into specific time periods.

More resources ​

• AI Gateway overview: Understand AI Gateway, its key components, and how it works.
• Configure AI Gateway: Add providers, create API keys, and define policies.