Introducing Gemini 2.5 Flash

We're excited to announce the public preview of Google’s Gemini 2.5 Flash across all our resources, including bots and agents. This enhancement combines speed, cost-efficiency, and adaptive problem-solving in a unique hybrid reasoning architecture.

Key features of Gemini 2.5 Flash:

1. Enhanced Multimodal and Long-Context Capabilities: The model supports 1 million tokens of context across various input types, making it valuable for tasks like legal contract review, medical record summarization, or video scene analysis.
2. Superior Performance and Cost Efficiency: Gemini 2.5 Flash outperforms comparable models on multi-step reasoning tasks, offering superior accuracy and lower inference costs.

Developers can immediately access Gemini 2.5 Flash through our SDK, API endpoints, bots, and agents. For advanced configurations, refer to our updated documentation.

We encourage customers of ChatBotKit to explore the model's potential in enhancing existing workflows and enabling new capabilities. As we move towards general availability, expect further optimizations and expanded tooling support.