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

This article announces the general availability of GLM 5.3 from Z.ai on Amazon Bedrock, a powerful model for agentic coding and long-horizon software engineering work.

• GLM 5.3 is a mixture-of-experts model with 753B total parameters and 40B active per token
• Supports 1-million-token context window with up to 128K output tokens
• Reasoning always enabled with selectable effort levels to balance latency and performance
• Includes explicit prompt caching to reduce latency and input costs
• Available to eligible enterprise customers via US and Global cross-Region inference profiles
• Accessible through Amazon Bedrock console and supported APIs

GLM 5.3 on Amazon Bedrock provides enterprise customers with a state-of-the-art open weight model backed by AWS security and compliance standards.