Google Launches Enhanced Gemini 1.5 API Models with Improved Performance and Lower Development Costs

Google has just launched two stable versions of its Gemini 1.5 API models for developers. These new models promise better performance and lower costs for building apps.

The tech giant introduced the Gemini 1.5 Pro (gemini-1.5-pro-002) and Gemini 1.5 Flash (gemini-1.5-flash-002) models on September 24.

These updated models are much improved compared to earlier version 001.

Google's Gemini 1.5 Models: Major Improvements, Price Cuts, and Enhanced Developer Tools

The new Gemini models show significant advancements in several areas. For example, they are better at generating code, doing math, solving problems, and analyzing videos. This means that developers can create apps that work more efficiently and effectively.

Secondly, Google made a significant change by cutting the price of its Gemini 1.5 Pro model by over 50%.Also, this model can handle three times more requests and has lower delays than the older experimental versions.

Google's notes highlighted some critical improvements in both Gemini 1.5 models. For one, they have become much better at providing accurate information and are less likely to create false or misleading responses, often referred to as hallucinations."

Additionally, these models are better at following instructions and understanding multiple languages, at least up to 102.They also excel at generating SQL code and can understand audio and documents more effectively."

Google has shortened the summarization lengths for both Gemini 1.5 models. This means that the models will now provide more concise summaries.

In addition, Google has offered suggestions to help chat-based product developers enhance their API's conversational abilities.

With these options, developers can create more engaging and interactive chat experiences.

Starting October 1, Google will cut prices for the Gemini 1.5 Pro API when handling prompts with fewer than 128,000 tokens.

The price of input tokens will be reduced by 64%, while the price of output tokens will drop by 52%. Additionally, the cost of incremental cached tokens will also be cut by 64%.

This means that using the API for smaller tasks will be much cheaper, which can help developers keep their costs down.

Google Boosts Gemini Models as OpenAI Introduces Advanced Voice Features in Competitive AI Landscape

Google is also making it easier for developers to work with Gemini by raising the rate limits for the paid tiers of both models.

The 1.5 Flash model's limit will increase to 2,000 requests per minute (RPM). This is an improvement from the previous limits of 1,000 RPM for Flash and 360 RPM for Pro.

As such, developers will find it easier to make more requests in a shorter amount of time, making it easier for them to build and improve their applications.

Google also introduced Gemini 1.5 Flash-8B, a smaller experimental version of the 1.5 Flash model.While its benchmark scores are lower than those of the full versions, this model still offers noticeable improvements in performance.

Developers can now access all versions of the Gemini 1.5 models on Google AI Studio and through the Gemini API. At the same time, Google's biggest AI competitor, OpenAI, has started rolling out a new feature called Advanced Voice" to select ChatGPT users.

This feature is designed to make AI communication faster and more natural, mimicking human conversation more closely.

Open AI has also introduced five new voices: Arbor, Maple, SXol, Spruce, and Vale.These will add to the existing voices, Breeze, Juniper, Cove, and Ember voices, offering more variety for users to choose from.

The post Google Launches Enhanced Gemini 1.5 API Models with Improved Performance and Lower Development Costs appeared first on The Tech Report.