Google Tools for Building Generative AI
Google Debuts PaLM API and MakerSuite for Generative AI App Developers
ERIC HAL SCHWARTZ
Google has introduced two new tools for building generative AI projects. The new PaLM API offers access to Google’s PPaLM large language model within their own applications, while the new MakerSuite app augments the API with a prototyping environment to test and improve on ideas, with an eye toward adding ways to engineer prompts, produce synthetic data, and customize the fine-tuning of a model. Both the PaLM API and MakerSuite are in Private Preview, with a waitlist opening soon.
PALM AND MAKERSUITE
The PaLM API creates an entryway for developers keen to access Google’s LLMs. PaLM is named for its origin in the Pathways Language Model. It’s used in a variety of projects, including teaching robots to understand orders, as well as other Google projects to foster communication between humans and AI. The API gives developers access to the relevant models, as well as the option to adjust and customize the models with their own data to meet their needs.
The MakerSuite setup is another way to link developers to Google’s generative AI models, but with a low-code approach that could entice a broader array of developers. Users can access two versions of PaLM, one for more chatbot-style interactive apps, and the other more for single input/output designs. Within MakerSuite, users can test out different inputs to see how the model responds, and fine-tune the prompts to better produce the kind of answers it wants. The improved version can then be translated into code to embed within the PaLM API and the software the developer is building with the API.
“We’re seeing a new wave of generative AI applications that are transforming the way people interact with technology – from games and dialog agents to creative brainstorming and coding tools. At Google, we want to continue making AI accessible by empowering all developers to start building the next generation of applications with generative AI by providing easy-to-use APIs and tools,” Google vice president of engineering Scott Huffman and senior director of product management Josh Woodward explained in a blog post. “The PaLM API is a simple entry point for Google’s large language models, which can be used for a variety of applications. It will provide developers access to models that are optimized for multi-turn use cases, such as content generation and chat, and general purpose models that are optimized for use cases such as summarization, classification, and more.” ... '
No comments:
Post a Comment