AWS enhances Amazon Bedrock with AgentCore runtime, prompt caching, and Kimi K3 model
AWS announced several enhancements to its Amazon Bedrock service, including a new AgentCore runtime for faster agent starts, prompt caching to optimize cost and latency, and the availability of the Kimi K3 large language model. These updates, reported by the AWS Machine Learning Blog, aim to improve performance, efficiency, and model choice for users. All features are presented as shipped and available for immediate use.
What every outlet reports
- AWS introduced new features for Amazon Bedrock
- A new AgentCore runtime was launched for agents
- Prompt caching was introduced for cost and latency optimization
- The Kimi K3 large language model is now available on Bedrock
All 3 articles
optimist1
Introducing Kimi K3 on Amazon BedrockoriginalAWS Machine Learning Blog · company statementpragmatist2
Optimizing cost and latency with Amazon Bedrock prompt cachingoriginalAWS Machine Learning Blog · company statementThe new AgentCore runtime: Elastic, optimized, and consistently fast startsoriginalAWS Machine Learning Blog · company statement