AWS adds single-shot and retrieval-only modes to Bedrock knowledge bases

AWS expanded AgenticRetrieveStream in Amazon Bedrock Managed Knowledge Base with retrieval-only and single-shot retrieval-augmented generation. Developers can now return passages, retrieve once and generate an answer, or use multi-step retrieval without switching APIs.

Applications choose a mode for each request by setting maxAgentIteration from zero through five and selecting whether to generate a response. The setting does not automatically route requests between modes.

At zero, the service performs one retrieval and returns passages without invoking a foundation model. Setting generateResponse=true with zero is invalid and returns a 4xx error, according to AWS. AWS describes this mode as functionally equivalent to calling Retrieve directly, although the AgenticRetrieveStream result omits the score and location fields included in a Retrieve response.

At one, the service performs one retrieval and one answer-generation call without a planning round or query rewrite. AWS says this can reduce latency and model usage compared with agentic retrieval, but the company published no workload-specific measurements for latency, usage, cost or answer quality.

Values from two through five retain the existing agentic behavior. With response generation enabled, a value of n permits as many as n - 1 planning rounds before the final generation call. AWS counts each foundation-model call as one iteration; an initial speculative retrieval runs before planning and does not count toward the limit.

All three modes can search multiple knowledge bases, and one request can specify up to five retrievers. The stream returns retrieval results and trace events. When generation is enabled, it also streams response text and includes the generated answer and citations in the final result.