#google-ai

Use Google's Gemini for multimodal AI. Capabilities: 1. Text and image input simultaneously. 2. Vision understanding for analysis. 3. Long context window (up to 1M tokens). 4. Function calling support. 5. Code generation and execution. 6. Gemini Pro vs Ultra models. 7. Streaming responses. 8. Safety settings configuration. Use for image captioning, OCR, and visual Q&A.