
4/17/2025 · Tulsee Doshi
What this post added
Introduced Gemini 2.5 Flash, a new model available via the Gemini API in Google AI Studio and Vertex AI. This model features enhanced reasoning capabilities, a hybrid reasoning approach with controllable 'thinking' processes, and a 'thinking budget' parameter for fine-grained control over quality, cost, and latency. It offers comparable performance to leading models at a lower cost and size, with options to turn thinking on or off, or set a specific token budget for the thinking phase.