google/gemini-2.5-flash

Google’s hybrid “thinking” AI model optimized for speed and cost-efficiency

10.2K runs
Playground API Examples README

Readme

Gemini 2.5 Flash

Gemini 2.5 Flash is Google DeepMind’s cost-efficient, high-speed multimodal model designed for production workloads.
It balances speed, reasoning, and controllable “thinking depth,” making it ideal for developers who need performance at scale.

Key Features

  • Multimodal Input: Supports text, images, audio, and video as inputs.
  • Long Context Handling: Works with extremely long inputs (up to ~1 million tokens).
  • Controllable Reasoning: Developers can choose how much internal reasoning (“thinking”) the model applies.
  • Optimized for Speed & Cost: Fast inference times with efficient compute usage.
  • Flexible Output: Generates text, captions, summaries, structured data, and more.