Google at its I/O 2024 conference introduced Gemini 1.5 Flash, a lighter-weight model built for scale. According to the company it is optimised for tasks where low latency and cost matter most. 1.5 Flash will be available in AI Studio and Vertex AI. Besides this, Google is also upgrading Gemini 1.5 Pro which is now being incorporated into several Google products, such as Workspace applications and Gemini Advanced.
The newest model in the Gemini family, 1.5 Flash, is also the fastest model provided by the API. It is said to be more cost-efficient to serve, optimised for high-volume, and optimised for large-scale, high-volume and high-frequency jobs. Compared to the 1.5 Pro, this model weighs less but is still capable of multimodal reasoning over large volumes of data. Gemini 1.5 Flash is Google’s smallest device-running model and is positioned between Gemini 1.5 Pro and Gemini 1.5 Nano.
Demis Hassabis, CEO of Google DeepMind said, “1.5 Flash excels at summarization, chat applications, image and video captioning, data extraction from long documents and tables, and more.”
1.5 Flash operates effectively for a variety of tasks, including summarization, chat applications, image and video captioning, data extraction from long documents and tables, and more due to 1.5 Pro’s “distillation” process, which transfers essential knowledge from a larger model to a smaller, more efficient model.
The Gemini 1.5 Flash isn’t intended for the public. Instead, developers can use Google-designed technology more quickly and less expensively to create their own AI products and services.
Also Read: Tata Play and Amazon Prime come together to offer new Binge plans
Apart from this Google has also improved 1.5 Pro ability to generate code, logical reasoning and planning, multi-turn conversation, and audio and image understanding through data and algorithmic advances. in addition to expanding its context window to 2 million tokens.
In addition to this 1.5 Pro can now carry out more intricate instructions, including ones that specify product-level behaviour involving role, format and style. As per the company, it has improved control over the model’s responses for specific use cases, like crafting the persona and response style of a chat agent or automating workflows through multiple function calls. Additionally, by configuring the system, users may now control how the model behaves.
Also Read: Vivo X100s, Vivo X100s Pro with MediaTek Dimensity 9300+ SoC launched
As a result of Google’s addition of audio understanding to the Gemini API and Google AI Studio, 1.5 Pro is now able to comprehend both audio and images for videos that are submitted to Google AI Studio. Moreover, Gemini 1.5 Pro is also being incorporated into several Google products, such as Workspace applications and Gemini Advanced.
