.svg)
New serverless model customization capability in Amazon SageMaker AI
New Serverless Model Customization Capability in Amazon SageMaker AI
AWS introduces a new serverless model customization capability in Amazon SageMaker AI, enabling AI developers to quickly customize popular models with supervised fine-tuning and advanced techniques like reinforcement learning. This fully managed service simplifies the end-to-end model customization workflow, from data preparation to evaluation and deployment, accelerating the process with an easy-to-use interface.
Developers can customize models such as Amazon Nova, Llama, Qwen, DeepSeek, and GPT-OSS with their own data. They can use supervised fine-tuning and the latest customization techniques, including reinforcement learning and direct preference optimization. Additionally, the AI agent-guided workflow (in preview) allows for generating synthetic data, analyzing data quality, and handling model training and evaluation—all serverless.
Available in the following AWS Regions: Europe (Ireland), US East (N. Virginia), Asia Pacific (Tokyo), and US West (Oregon). To join the waitlist for the AI agent-guided workflow, visit the sign-up page.
What to do
- Explore the SageMaker AI model customization page for more details.
- Visit the blog for the latest updates.
- Join the waitlist for the AI agent-guided workflow.
Source: AWS release notes
If you need further guidance on AWS, our experts are available at AWS@westloop.io. You may also reach us by submitting the Contact Us form.
