This webinar discusses large language models (LLMs) fine-tuning and training across different scale and various options. This includes fine-tuning/pre-training proprietary models with Bedrock and tuning/training open models with SageMaker AI training job which can leverage spot instances, SageMaker AI HyperPod with Slurm, and EKS. Learn also about larger scale training leveraging multiple GPUs and even multiple nodes with Ray on EKS or SageMaker AI HyperPod.

In addition to the training orchestration, this webinar will also highlight the available infrastructure, including the available GPUs on AWS, the way to run them on spot to save training cost, and also the supporting components (e.g. training observability, scalability,

Join us online to learn about the topics above, and discover potential AWS credits you could leverage to potentially bring your model into AWS Marketplace and SageMaker Jumpstart, bring your solution closer to your customers.