IBM expands AI capabilities with launch of Meta Llama 3 on Watsonx platform
IBM has announced the integration of Meta’s next-generation open large language model, Meta Llama 3, into its Watsonx AI and data platform. This significant update expands IBM’s model library, which already includes the Granite series and other models from leading providers, enhancing the platform’s capabilities to foster innovation across enterprises.
The addition of Llama 3 strengthens the ongoing collaboration between IBM and Meta, aimed at promoting open innovation in AI. This partnership previously gave rise to the AI Alliance, initiated late last year, which now boasts over 80 members from various sectors including industry, academia, and government. This alliance underscores the commitment of both IBM and Meta to lead a robust ecosystem of AI developers and researchers.
IBM Consulting and its Client Engineering teams have been actively deploying Llama models across numerous enterprise pilots, demonstrating the practical utility of these AI tools in real-world scenarios. A notable application includes the development of a content engine for the Recording Academy, which utilized Llama 2 to generate digital content that aligns with the established brand voice and standards of the GRAMMYs.
Meta has equipped Llama 3 with advanced capabilities, including models with 8 billion and 70 billion parameters, catering to a wide range of AI tasks such as summarization, classification, and more sophisticated needs like conversational AI and content creation. The 8B model is optimized for rapid training and deployment on edge devices, while the 70B model targets more demanding enterprise applications. Meta reports that Llama 3 shows improved performance over its predecessor, with further enhancements and new model sizes expected in the near future.
IBM now also hosts Code Llama 34B, a model specifically designed for code generation and translation, on the Watsonx platform.
This addition underscores IBM’s commitment to providing flexible AI solutions that can be deployed either as Software-as-a-Service (SaaS) or on-premises, allowing enterprises to scale AI applications in accordance with their specific data privacy and regulatory needs.
IBM’s update to its Watsonx platform with Meta Llama 3 represents a significant advancement in the utility and flexibility of AI technologies for enterprise applications. This strategic move not only bolsters IBM’s position in the AI landscape but also enhances its offerings to clients seeking to integrate and scale AI solutions within their operations. As AI continues to evolve, such partnerships and innovations will be crucial in shaping the future of how businesses leverage machine learning and data analytics for competitive advantage.