GLM 5.3, featuring a mixture-of-experts architecture with 753 billion total parameters and a 1-million-token context window, became generally available on Amazon Bedrock, providing powerful natural language reasoning capabilities with selectable effort levels to balance latency and performance.
This iteration builds upon GLM 5.2's base with scaled post-training improvements and includes explicit prompt caching to reduce latency and costs on repeated context use. Available to eligible enterprise customers, it is accessible across multiple AWS regions via the Bedrock console or APIs, supported with AWS’s security and compliance standards.
Source & further reading
GLM 5.3 by Z.ai is now generally available on Amazon Bedrock ↗Source published 2026-10-05. This report was prepared with AI assistance from the linked primary-source material. Claims and limitations should be checked against the original announcement or research.