Changelog
added

Added Gemini Flash 1.5 Model Support & Deprecated Fine-tuned Models

We've added support for Gemini Flash 1.5 as a new model option while deprecating fine-tuned models across all endpoints. This update brings ultra-fast generation capabilities to our API while streamlining our model offerings.

New Model Details

The Gemini Flash 1.5 model offers:

  • 1-3 second response latency
  • High-quality output comparable to base models
  • Optimized for high-volume requests and real-time applications

Example usage in generate-content

{
  "model": "gemini_flash_1.5"
}

Important Deprecation Notice

All fine-tuned model endpoints have been deprecated. Requests to these endpoints will now return an error. Please migrate your applications to use our base models instead.

Availability

This update is available on all active endpoints.