Govern Gemini models and Gemini CLI¶

WSO2 API Platform works alongside the Google products and services you already use. Govern Gemini model traffic, route Gemini CLI through AI Gateway, and send gateway telemetry to Google Cloud.

End-to-end guides¶

A step-by-step guide for Gemini CLI.

Configure Google Gemini CLI with AI Gateway

Route Gemini CLI requests through a Gemini provider and an App LLM Proxy.

Guide · Gemini CLI · Hybrid

AI Gateway with Gemini models¶

Route, secure, and govern traffic to Gemini models through WSO2 AI Gateway, including traffic from Gemini CLI.

Gemini as an LLM provider¶

Register the Gemini API with the gateway. Applications call the gateway and never see the Gemini API key.

Configure an LLM provider AI Workspace
Call Gemini through the gateway AI Workspace

Gemini provider AI Gateway
Configure an LLM provider AI Workspace

OpenAI-format requests¶

Keep clients on the OpenAI request format. A transformer policy converts requests and responses for Gemini.

OpenAI-format compatibility AI Gateway
OpenAI to Gemini transformer policy Policy Hub

Routing, token limits, cost control, and the other AI Workspace policies work the same way for every provider. See the AI Workspace overview.

Routing, token limits, cost control, and the gateway's other policies work the same way for every provider. See the AI Gateway documentation.

Google Cloud Monitoring and Cloud Trace¶

Send the metrics and traces of a gateway you run to Google Cloud through an OpenTelemetry Collector.

Metrics and traces¶

Point the collector at Cloud Monitoring for metrics and at Cloud Trace for traces.

Send AI Gateway traces to Cloud Trace AI Gateway

Send API Gateway metrics to Cloud Monitoring API Gateway
Send API Gateway traces to Cloud Trace API Gateway
Send AI Gateway traces to Cloud Trace AI Gateway

Architecture and responsibility model¶

WSO2 API Platform and Google split the work like this:

Scenario	WSO2 role	Google role	Runtime traffic
Gemini CLI through AI Gateway	Authenticates developers and applies guardrails, rate limits, and analytics	Hosts the Gemini models and bills the API key	Goes through WSO2 AI Gateway to the Gemini API
Applications calling Gemini models	Proxies and governs LLM requests	Hosts the Gemini models	Goes through WSO2 AI Gateway to the Gemini API
Gateway telemetry in Google Cloud	Emits metrics and traces from the gateway	Stores and displays the telemetry	The collector sends telemetry to Google Cloud without affecting API traffic

Learn more¶

Related pages in the Google documentation:

To browse every guide by goal, see the Guides overview.