Google that last week during the cloud Next eve of the new model 2.5 Flash It has now released the preview of this argume through the Gemini API on Google AI Studio and Vertex AI platforms. Jamina 4.1 Flash is Google’s first hybrid artificial ielligence; This means that it is possible to enable or deactivate the reasoning.
The new Gemini 2.5 Flash model coinues its previous version, Gemini 2.0 Flash, and has been iroduced with remarks on its reasoning abilities. Google claims this model has a very good performance ratio. In addition to developers, Jina’s users can access this new version and use it in the Canvas environme to optimize documes and codes.
Cost of using Google’s Jina Model 2.5
The details of the new Google model pricing are as follows:
- Input fee: $ 0.15 per token
- No argume output: $ 0.60 per token
- Output with reasoning: $ 3.50 per token
It is noteworthy that Jina 2.5 Flash is the first fully fully hybrid model of Google in which it is possible to enable or deactivate the reasoning. This feature helps developers optimize their responses based on target quality, cost and delay.
According to the published benchmarks, the Gemini 2.5 Flash, despite its low cost, performs acceptable competing models such as Ahropic and Grak models. Of course, the O4-Mini model, recely released by Openai, performs better than the Gemini 2.5 Flash preview, but it costs much higher.
