According to @CNBC, IBM’s Arvind Krishna said generative AI augments, not disrupts, the software unit, highlighting hybrid cloud and watsonx growth.

According to AndrewYNg, a new Red Hat-backed course shows how vLLM and quantization cut memory and cost for high-concurrency LLM serving.

According to DeepLearningAI, a free course with Red Hat teaches vLLM serving, LLM quantization, and benchmarking for speed, cost, and accuracy.