Gemini API

Gemini API is Google AI's developer interface (ai.google.dev) that provides programmatic access to the multimodal Gemini model family developed by Google DeepMind. It lets developers integrate Gemini capabilities — text generation, image, audio, video and code understanding — into applications via REST and SDKs.

The API exposes successive generations of Gemini models, including Gemini 3 Pro, Gemini 3 Flash, Gemini 3.1 Pro, Gemini 3.1 Flash-Lite, Gemini 3.1 Deep Think and image-generation models (Nano Banana / Gemini 3.1 Flash Image). It supports synchronous and streaming responses, function calling, tool use, structured output and long-context inputs (up to 1M tokens in selected models).

Pricing is usage-based on tokens (input / output / cached input) with a free tier offering limited quotas and a paid tier with higher limits. SDKs are available for Python, JavaScript/TypeScript, Go and Java through the Google Gen AI SDK. The same model set is also exposed via Vertex AI on Google Cloud, in an enterprise variant with additional security and regional guarantees.