Gemini 3.5 Flash (ai.google.dev)

The launch of Gemini 3.5 Flash introduces a cutting-edge AI model designed to enhance performance in real-world applications, boasting significant improvements in speed and cost-efficiency. This model is particularly geared toward facilitating complex workflows and extending the capabilities of sub-agents, making it suitable for multi-step, long-horizon tasks. Its strength lies in executing swift agentic loops, which are invaluable for intricate coding tasks that require rapid iterations. Gemini 3.5 Flash supports a wide range of input data types—including text, images, video, audio, and PDFs—and features a staggering input token limit of 1,048,576, with an output token limit of 65,536. Key functionalities include batch API support, function calling, and enhanced grounding capabilities with Google Maps. This makes the model not only highly versatile but also positioned to handle complex queries and operations effectively. For the AI/ML community, this advancement signifies a substantial leap toward creating more efficient, scalable solutions that can significantly impact industries relying on intelligent automation and data processing.