Gemini 3.5 Flash Lite (batch)

Gemini 3.5 Flash-Lite (Batch) is a lightweight AI model developed by Google DeepMind, designed for speed, cost efficiency, and large-scale AI deployment. Unlike flagship Gemini models focused on maximum reasoning performance, Flash-Lite is optimized for high-volume workloads and fast execution, making it ideal for batch data processing, document analysis, information extraction, AI Agent sub-tasks, and automated workflows. The Batch version supports asynchronous batch inference, allowing businesses to process large numbers of AI requests efficiently while reducing operational costs. Gemini 3.5 Flash-Lite supports multimodal inputs, including text, images, videos, audio, and PDF documents, while providing advanced developer capabilities such as Function Calling, Structured Output, Code Execution, File Search, Search Grounding, and URL Context understanding. With a 1,048,576-token input context window (approximately 1 million tokens) and support for up to 65,536 output tokens, the model can handle large documents, enterprise knowledge bases, technical resources, and complex information extraction tasks. Gemini 3.5 Flash-Lite is designed to deliver advanced AI capabilities at a lower cost, making it suitable for scalable business applications, automation systems, and AI Agent workflows.