Google has publicly released Gemini 1.5 Flash-8B, the newest version of the company’s Flash creation and production model that is ready for developers. This new model is claimed to be 50% less expensive than the old model, the Gemini 1.5 Flash.
Gemini 1.5 Flash-8B Simple To Use
Google goes on to explain Flash-8B as a micro version of the Gemini 1.5 Flash model with only a slight difference in the total scores on benchmark tests. It does very well in cognate work such as chat, transcription services, and, particularly, long context language translation. Google has fine-tuned the Gemini 1.5 Flash-8B for multiple use cases such as chat, transcription, and long-context language translation so that it performs nearly as well as the more heavy-duty Gemini 1.5 Flash version.
This level of efficiency, however, does not have fatalities on quality or on speed since the model can handle 4000 requests per minute. These improvements have placed the Gemini 1.5 Flash-8B among the best bet instruments for developers who are looking for reasonable yet efficient AI solutions.
In this stable release, Google DeepMind is also focusing on the affordable nature of its products. The new pricing model is charged at $0.0375 per million input tokens and $0.15 per million output tokens of the prompt below 128,000 characters. For cached prompts, the fee is $0.01 per million tokens.
Gemini 1.5 Flash-8B is open to anyone and can be acquired in AI Studio by Google and accessed by the Gemini API. For the next update, it shall only be open for those on the paid version of the platform, with billing for this new model starting on October 14th.