Models in Gemini API Docs
Summary
Documentation listing all AI models available through the Gemini API, including their capabilities and endpoints.
Key quotes
"Gemini 3.8 Flash is our most intelligent Flash model for long-horizon software engineering and complex workflows."
"Nano Banana 2.1 delivers high-efficiency image generation with improved visual quality and multi-turn consistency."
"Gemini 3.5 Transcribe offers low-latency speech-to-text with language detection, diarization, and timestamps."
"Preview models may be used for production but have restrictive rate limits and imply eventual deprecation."
This documentation page provides an overview of the range of models available via the Gemini API, including text, image, audio, video, and specialized models like robotics and embedding tools. Models are categorized by type (e.g., Flash, Pro, Nano Banana) and use case, with endpoints specified for integration. Updates such as the launch of Gemini 3.8 Flash and Nano Banana 2.1 are highlighted, alongside legacy model deprecations.