Gemini-1.5-Pro-002 and Flash-002 AI models are now available
Google on Tuesday released newer versions of its Gemini 1.5 Pro artificial intelligence (AI) model. These come just months after the Mountain View-based tech giant released the latest version of Gemini, which increased the context window to 2 million tokens. The company said that these AI models, dubbed Gemini-1.5-Pro-002 and Gemini-1.5-Flash-002, will not only offer higher output and lower costs, but will also offer a higher speed limit for users. Additionally, filter settings have also been updated to make the AI models adhere more closely to instructions.
New Gemini AI models released
In a blog postThe company detailed the Gemini-1.5-Pro-002 and Gemini-1.5-Flash-002 AI models. These models are currently available as experimental model releases and are built on Gemini 1.5 Pro, which was first released at Google I/O in May. They are currently available to developers and enterprise customers of the company. Developers can access them for free via Google AI Studio and the Gemini API. Enterprises can access them via Vertex AI.
The latest Gemini 1.5 Pro and Flash models also outperformed the previous-generation Gemini model, according to internal testing conducted by Google. The company claimed that the new models saw a seven percent increase in the Massive Multitask Language Understanding Pro (MMLU-Pro) benchmark. Further, the AI models were said to offer around 20 percent improvement on the MATH and HiddenMath benchmarks compared to the Gemini 1.5 Pro.
Gemini-1.5-Pro-002 and Gemini-1.5-Flash-002 AI models also offer increased speed capping. Speed caps are the daily usage limits for users. With the 1.5 Flash model, users get 2,000 requests per minute (RPM) and the 1.5 Pro model offers 1,000 RPM. Google said these limits are being increased to allow developers to build with the new versions of Gemini.
It’s not just the speed limit that’s getting an upgrade. With the Gemini-1.5-Pro-002 and Flash-002, the company has also increased the output tokens per second, making the models more responsive and faster at generating long blocks of text.
A major upgrade with these AI models is the improvement to filters. Google said that the new Gemini AI models will better adhere to prompts and follow instructions thanks to these updated filters. Google is also improving its suite of safety measures to ensure that the AI models don’t generate anything malicious. Notably, the default filters will not be applied in new AI models, allowing developers to choose the configuration they prefer.
