Four Bedrock request-level settings drive inference cost differences

An AWS Machine Learning Blog post outlines four request-level choices on Amazon Bedrock — cache duration, service tier, reasoning effort, and maxTokens — that change inference costs even after prompt and application tuning.

Read the full story at AWS Machine Learning Blog ↗

Timeline · 1 report

2026-11-12 20:51 · AWS Machine Learning Blog
Four Amazon Bedrock inference choices that affect cost

More stories

AWS outlines centralized data lake for Amazon Bedrock tokenomics observability — AWS Machine Learning Blog
AWS recaps October 2026 updates for Bedrock, AgentCore, and Strands — AWS Machine Learning Blog
Anthropic prepares for IPO with Amazon, Google, and Nvidia as major backers — CNBC Technology
AWS Emulator: AI Agents Build Local Copy of 200+ AWS Services — AWS Machine Learning Blog
AWS adds fine-grained access control and migration tooling for Amazon Bedrock Managed Knowledge Bases — AWS Machine Learning Blog
Meta infrastructure chief Santosh Janardhan to step down in early 2027 — DigiTimes Asia
AWS launches Web Search API for Bedrock AgentCore as adoption cases emerge — AWS Machine Learning Blog
AWS adds usage-based pricing for SageMaker AI bidirectional streaming — AWS Machine Learning Blog

Get the daily brief of stories like this at 6:30 every morning →