GLM 5.3 by Z.ai is now generally available on Amazon Bedrock

Amazon Bedrock has introduced support for GLM 5.3, developed by Z.ai, expanding the platform’s model options for enterprise customers. The model is designed for agentic coding and long-horizon software engineering tasks, providing a mixture-of-experts architecture with 753 billion total parameters and approximately 40 billion active parameters per token. It builds on the base of GLM 5.2, incorporating improvements from scaled post-training to enhance performance and capabilities.

GLM 5.3 features a 1-million-token context window and supports up to 128,000 output tokens, enabling processing of extensive codebases or documentation. Reasoning is enabled by default with selectable effort levels, allowing users to balance latency, token consumption, and task performance according to their needs. The model also includes explicit prompt caching, with cache points on system prompts and messages to reduce latency and input costs during repeated model interactions.

Access to GLM 5.3 on Amazon Bedrock is available to eligible enterprise customers, with support for US and Global cross-Region inference profiles. Users can deploy the model through the Amazon Bedrock console or programmatically via supported APIs, ensuring flexibility in integration and deployment methods.

For detailed information on supported AWS Regions, endpoints, APIs, features, inference profiles, and pricing, customers are directed to the Amazon Bedrock documentation. Additional resources, including a launch blog post, provide further insights into the model’s capabilities and use cases.