Computing giant IBM has recently announced that it can develop a Deepseek-R1 distillation version in Watsonx.ai, an enterprise-grade AI development studio that promotes the creation, training and development of AI models in business applications. Developers can upload and deploy custom foundation models to use Watsonx.ai inference functions.
The company has also outlined a detailed steps to support users to develop Deepseek-R1 distilled variants for inference.
“The release of Deepseek-R1, the same optional model for O1 series of O1 series, is an important step in this direction,” said IBM. “I hope other model providers will follow the lawsuit and encourage them to contribute to the growth of the open source AI ecosystem.”
On Wednesday, IBM announced the fourth quarter of 2024. The company has reported a $ 17.6 billion (up 1 %), but has increased software revenue by 10 %. The total profit margin of IBM increased by 40 basic points to 59.5 %. However, consulting revenue decreased by 2 %, and infrastructure revenue decreased by 8 %.
IBM’s CEO Arvind Krishna states: He also revealed that the company’s AI products, including IBM concerts and other AI assistants, are leading about 8 points.
The company also expects at least 5 % of revenue growth and a free cash flow of about $ 13.5 billion in 2025.
“We are convinced that we are convinced that we are trying to accelerate consulting growth as we move forward with partnerships and skills, and Genai’s early leadership.
Last November, the company released a new variant of the open source AI model GRANITE 3.0. Armand Ruiz, Vice President of the IBM AI Platform, has released a dataset where models are trained. This is a practice that IBM complied with even in the past with the release of a new model.
“This is true transparency. Other LLM providers do not share such detailed information about the training dataset,” said Ruiz.
