Watsonx Updates, Open-Source LLMs Unveiled at IBM Think

IBM THINK — IBM will announce multiple new updates to the watsonx platform and other AI-powered tools Tuesday during IBM Think 2024, its annual event being held this year in Boston. Channel Futures is there.

First, IBM will release a number of Granite models, its collection of large language models based on decoder-only transformer architecture, as open source. This open approach to the technology will allow clients, developers and experts to build on what IBM's developed so far and push it well past its boundaries.

IBM's Arvind Krishna

“We firmly believe in bringing open innovation to AI. We want to use the power of open source to do with AI what was successfully done with Linux and OpenShift,” said IBM CEO Arvind Krishna. “Open means choice. Open means more eyes on the code, more minds on the problems, and more hands on the solutions. For any technology to gain velocity and become ubiquitous, you’ve got to balance three things: competition, innovation, and safety. Open source is a great way to achieve all three.”

The new Granite models will be available on collaborative platforms HuggingFace and Github as of Tuesday. The vendor upheld the Granite models as stronger then its competitors, claiming that they are able to outperform larger general-purpose LLM when involved in code-related tasks.

Related:IBM Plans MSP Program, Honors Partner Plus Award Winners

AI Tools Unveiled at IBM Think 2024

The vendor also said at IBM Think 2024 it will launch InstructLab in conjunction with RedHat. InstructLab will allow developers to build models specific to their business domains or industries using just their data so they can see what AI can do for them rather than what the model providers value.

IBM further unveiled several upcoming updates and enhancements to its family of watsonx assistants. These include an assistant designed to assist with enterprise Java applications, an Assistant for Z that will transform how users interact with the system and an expansion of the Code Assistant for Z to help clients grasp and understand how the models work.

IBM also will expand its Nvidia GPU offerings to include GPU L4 and L40 models, as well as support for Red Hat Enterprise Linux AI and OpenShift AI.

Moreover, the company intends to launch several new generative AI products to help organizations observe, govern and optimize their data sources for AI workloads. These include IBM Data Product Hub, Data Gate for watsonx and several updates to watsonx's data.

Tuesday's AI news followed Monday's IBM Partner Plus Day, where the company announced its new MSP program and unveiled the recipients of its Partners Plus awards. IBM also expanded its partnership with AWS cloud migration partner Innovative Solutions to bring additional generative AI tools to cloud-based applications and the businesses using them.