Last week, Google introduced its latest and most powerful generative AI model, Gemini, along with a detailed roadmap and immediate availability for users to explore it through Bard. Today, Google announced the availability of Gemini Pro to developers and organisations via the Gemini API on Google AI Studio, while enterprises can access it through Google Cloud’s Vertex AI platform.
Gemini Pro outperformed GPT-3.5 in several benchmarks, including MMLU and GSM8K, which measure language understanding and grade school math reasoning, respectively. Developers will have free access to Gemini Pro and Gemini Pro Vision, a multimodal endpoint that accepts text and image input, through Google AI Studio. Vertex AI customers will also have access to the models at no cost until general availability next year.
In addition to Gemini Pro, Google is introducing Imagen 2, an upgraded image AI model that delivers improved image quality and text rendering in multiple languages. Imagen 2 can generate a range of creative and realistic logos for businesses. Google is also launching MedLM, a family of AI models tailored for the healthcare industry, and is expanding the capabilities of Duet AI in Google Cloud, its AI-powered collaboration solution.
Must Read: AI Images consume as much energy as charging your smartphone
In addition to Gemini Pro, Google is rolling out a range of other AI tools and models to enhance the capabilities of developers and enterprises. Here’s a breakdown of the key announcements:
Google Gemini Pro API in AI Studio and Vertex AI:
- Gemini Pro is now accessible to developers through the Gemini API on Google AI Studio.
- Enterprises can also leverage Gemini Pro via Google Cloud’s Vertex AI platform for additional customization and Google Cloud features.
- Developers can easily transition their AI Studio code to Vertex AI when ready for a fully managed AI platform.
- Developers can get started by visiting ai.google.dev.
Also read:Beginning of Google’s Gemini Era: 10 amazing things Gemini can do
Google Gemini Pro Introducing Imagen 2 on Vertex AI:
- Google has updated Vertex AI with Imagen 2, the most advanced text-to-image diffusion technology from Google DeepMind.
- Imagen 2 offers significantly improved image quality and advanced features, including the generation of creative and realistic logos for businesses, brands, and products.
- It excels at rendering text in multiple languages, addressing challenges often faced by text-to-image tools.
Google Gemini Pro MedLM in Vertex AI:
- Google introduces MedLM, a family of foundation models fine-tuned specifically for the healthcare industry.
- Accessible (via allow list) to Google Cloud customers in the U.S. through Vertex AI, with plans to include it in the open Model Garden in the coming weeks.
- Building on the Med-PaLM 2 foundation model, MedLM will soon incorporate Gemini-based models to offer enhanced capabilities.
Must Read: Gemini Pro VS Gemini Ultra: What are the key differences?
Google Gemini Pro Duet AI for Developers:
- Duet AI for Developers is now generally available, serving as an always-on collaborator from Google Cloud.
- Provides AI-powered code and chat assistance within users’ preferred code editor and software development lifecycle tools.
- Streamlines running applications on Google Cloud, with built-in support for privacy, security, and compliance requirements.
- Google plans to integrate Gemini across the Duet AI portfolio in the coming weeks.
Google Gemini Pro Duet AI in Security Operations:
- Google Cloud becomes the first major cloud provider to make generative AI available to defenders on a unified security operations (SecOps) platform.
- Duet AI in Security Operations empowers defenders with advanced AI capabilities to enhance security measures.
As Google makes these innovative tools and models available, developers and enterprises are encouraged to explore the possibilities that Gemini and the expanded AI toolkit offer. The incorporation of Gemini across various platforms signals a commitment to advancing AI capabilities and opens up new avenues for creativity and problem-solving. Stay tuned for further developments as Google continues to push the boundaries of AI technology.
Must Read: Difference Between Gemini vs Bard vs ChatGPT vs Copilot vs Grok vs Ernie