- Machine Learning›
- Amazon SageMaker AI›
- Model customization
Model Customization with Amazon SageMaker AI
Accelerate AI model customization from months to days with serverless reinforcement learning and an AI agent-guided workflow
Why SageMaker AI for model customization
Amazon SageMaker AI empowers AI developers to customize popular models, such as Amazon Nova, Llama, Qwen, DeepSeek, and GPT-OSS, with the latest techniques like reinforcement learning in days. You can use the easy-to-use interface or AI agent-guided workflow (in preview) to quickly specify requirements, generate synthetic data, analyze data quality, and evaluate models for accuracy—all entirely serverless so you can focus on innovation rather than managing infrastructure.
Benefits
Fast track model customization with maximum accuracy
With an easy-to-use interface or an AI agent-guided workflow (in preview), you can complete the end-to-end model customization workflow, from data preparation to deployment, and accelerate the process from months to days.
Access the broadest set of customization techniques including reinforcement learning from AI feedback (RLAIF) and verifiable rewards (RLVR), supervised fine tuning (SFT), and direct preference optimization (DPO), all through an easy-to-use interface with built-in best practices or AI agent-guided workflow (in preview).
Quickly define your model customization use case in natural language and an AI agent builds a specification for you. The AI agent helps you generate synthetic data, analyze data quality, fine-tune models, and evaluate their performance based on your use case and success tenets.
Focus on model development instead of infrastructure management with end-to-end model customization that is completely serverless. SageMaker AI automatically handles compute provisioning, scaling, and optimization so you don’t have to.
Model customization made simple
Comprehensive capabilities to customize models across the end-to-end workflow
Synthetic data generation (in preview)
If real-world data is limited, you can easily generate synthetic data. If needed, the AI agent in SageMaker AI generates datasets based on data samples and contextual documents in the required format and structure for your selected model customization technique.
Advanced customization techniques
SageMaker AI supports the latest model customization techniques including supervised fine-tuning (SFT), direct preference optimization (DPO), and reinforcement learning from AI feedback (RLAIF) and verifiable rewards (RLVR).
End-to-end serverless model customization
SageMaker AI automatically selects and provisions the appropriate compute resources based on the model and data size—all without requiring you to select and manage instances.
Inference
Once you have achieved your desired accuracy and performance goals, you can deploy models to production in a few clicks to either SageMaker AI inference endpoints or Amazon Bedrock for serverless inference
LLMOps
You can automatically log all critical experiment metrics all without provisioning a tracking server or modifying code. Integration with MLflow also provides rich visualizations and an ingress into the MLflow user interface for further analysis.
Use cases
Craft your AI model to sound like your company’s voice and tone that consistently matches your style in every response.
Train your AI model to generate responses that users prefer. Collect feedback on multiple response options and optimize the model to consistently produce the most favored outputs.
Turn your AI model into an expert in your industry. Feed it industry knowledge so it understands your jargon, requirements, and best practices.
Easy-to-use interface
For more control and flexibility, you can use the easy-to-use interface to identify model evaluation criteria, select a model and customization technique, determine data enhancement needs, and deploy models.
AI agent-guided workflow (in preview)
Communicate your use case in natural language. An AI agent generates a specification that includes dataset guidelines, evaluation criteria, associated metrics, and customization technique.
If there are any dataset gaps, the AI agent helps generate synthetic data. You can approve the specification or continue the conversation with the agent to further refine the use case specification before launching model training.
Customers
Why our customers choose Amazon SageMaker AI for model customization
Collinear AI
"At Collinear, we build curated datasets and simulation environments for frontier AI labs and Fortune 500 enterprises to improve their models. Fine-tuning AI models is critical to creating high-fidelity simulations, and it used to require stitching together different systems for training, evaluation, and deployment. Now with Amazon SageMaker AI's new serverless model customization capability, we have a unified way that empowers us to cut our experimentation cycles from weeks to days. This end-to-end serverless tooling helps us focus on what matters: building better training data and simulations for our customers, not maintaining infrastructure or juggling disparate platforms."
Soumyadeep Bakshi, Co-founder, Collinear AI
Robin AI
"At Robin, we're redefining the role of legal in modern business and using AI to drive better decisions, faster actions, and sustainable growth. To provide our clients with better decision making, it is crucial that our AI models match how lawyers write contracts—from the specific format, tone, and preferences of individual lawyers. Previously, customizing models with proprietary data was a cumbersome process that was prone to errors. Now with the new serverless model customization capability in Amazon SageMaker AI, we can rapidly experiment with advanced techniques like reinforcement learning with verifiable rewards in just days. In addition, we're excited to try the AI agent-guided workflow so we can compare and verify our assumptions to help lawyers around the globe make better decisions faster.“
Diana Mincu - Director of Research, Robin AI
Resources
Did you find what you were looking for today?
Let us know so we can improve the quality of the content on our pages