The models—DeepSeek V3.2, MiniMax M2.1, GLM 4.7, GLM 4.7 Flash, Kimi K2.5, and Qwen3 Coder Next—cover frontier-level reasoning and agentic intelligence as well as autonomous and cost-efficient coding tasks. They offer users frontier-class performance with lower inference costs.
Powered by Project Mantle, Amazon’s distributed inference engine, these models support seamless onboarding, serverless inference, and enhanced quality-of-service features. They conform to OpenAI API standards, simplifying integration for enterprise AI applications spanning reasoning, coding, and lightweight production deployments.
Source & further reading
Amazon Bedrock adds support for six fully-managed open weights models ↗Source published 2026-02-10. 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.