Strands
A lightweight open-source agent orchestration framework from AWS that adopts a model-driven design philosophy. Developers declare tools and constraints, after which the underlying model autonomously determines the invocation order and reasoning path, making it suitable for multi-agent scenarios requiring dynamic decision-making.
Core Facts
Timeline (last 90 days)
Using Strands' BedrockModel to connect to Bedrock requires only a small amount of code, whereas SageMaker requires three API calls—create-model, create-endpoint-config, and create-endpoint—and in testing the endpoint took about 9 minutes to deploy to InService.
A Strands user reported that a turn with 18 tool calls resulted in 0 cache reads and 134,082 tokens of cache writes.
Strands' design philosophy is model-driven: developers only need to declare tools and constraints, and the underlying model autonomously determines the invocation order and reasoning path, rather than hardcoding execution steps.
Compared to frameworks such as LangGraph and AutoGen, Strands operates at a higher level of abstraction with less code, but relies more heavily on the reasoning capabilities of the underlying model.
The original article uses a supply chain decision-making scenario as an example to build a multi-agent system with the Strands framework, with Strands responsible for defining each agent's role, the set of callable tools, and the collaboration process.
Amazon Bedrock AgentCore is a managed service launched by AWS in 2025, designed specifically for running and orchestrating AI agents in production environments.
All Facts (6)
Amazon Bedrock AgentCore is a managed service launched by AWS in 2025, designed specifically for running and orchestrating AI agents in production environments.
80%UnverifiedUsing Strands' BedrockModel to connect to Bedrock requires only a small amount of code, whereas SageMaker requires three API calls—create-model, create-endpoint-config, and create-endpoint—and in testing the endpoint took about 9 minutes to deploy to InService.
50%UnverifiedStrands' design philosophy is model-driven: developers only need to declare tools and constraints, and the underlying model autonomously determines the invocation order and reasoning path, rather than hardcoding execution steps.
50%UnverifiedCompared to frameworks such as LangGraph and AutoGen, Strands operates at a higher level of abstraction with less code, but relies more heavily on the reasoning capabilities of the underlying model.
50%UnverifiedThe original article uses a supply chain decision-making scenario as an example to build a multi-agent system with the Strands framework, with Strands responsible for defining each agent's role, the set of callable tools, and the collaboration process.
50%UnverifiedA Strands user reported that a turn with 18 tool calls resulted in 0 cache reads and 134,082 tokens of cache writes.
50%