gemini-api - Gemini API on Agent Platform with Google Gen AI SDK
Guide usage of the Gemini API on Agent Platform with the Google Gen AI SDK across Python, JS/TS, Go, Java, and C#.
Tags
Updated: 2026-06-30Capabilities
Typical Inputs
Typical Outputs
What this skill does
- generate text content
- process multimodal inputs
- call user-defined functions
- generate structured JSON output
- cache large contexts
- generate text embeddings
- stream real-time audio and video
- run batch predictions
- generate and edit images
- detect objects in images and video
- tune models
- adjust safety filters
Inputs
- Google Cloud credentials
- Google Cloud project ID
- User input text or media
- JSON schema for structured output
- Fine-tuning datasets
Outputs
- Generated text responses
- Generated images
- Generated embeddings
- Structured JSON responses
- Streamed audio and video responses
- Batch prediction results
- Cached context
- Fine-tuned model
Requirements
- Active Google Cloud credentials
- Agent Platform API enabled
- Google Gen AI SDK installed
- Supported programming language runtime (Python, JS/TS, Go, Java, C#)