Overview
- Gemini 2.5 Pro and Flash reach general availability in stable releases, enabling production deployments.
- Gemini 2.5 Flash-Lite enters preview as a high-throughput, low-latency model with reasoning off by default to optimize cost and speed.
- Pricing for Flash has been revised to $0.30 per million input tokens and $2.50 per million output tokens, with a single tier regardless of thinking mode.
- All Gemini 2.5 models feature dynamic reasoning capabilities that enhance accuracy across coding, math, science and multimodal benchmarks.
- Flash-Lite preview is accessible in Google AI Studio and Vertex AI and supports native tools such as Google Search grounding, code execution and URL context.