---
format: "aidr-story-markdown/v1"
id: "7f38ef45be453fbdbcd9655846c736a9e09a61c9aebabd31481e58fbb9000181"
canonical_url: "https://aidr.today/7f38ef45?lang=en"
title: "Use prompt caching with open-weight models on Amazon Bedrock"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-16T15:42:41.000Z"
category: "Tools"
topics: ["amazon","inference","devtools"]
source_urls: ["https://aws.amazon.com/blogs/machine-learning/use-prompt-caching-with-open-weight-models-on-amazon-bedrock/"]
summary: "Prompt caching allows open-weight models on Amazon Bedrock to reuse long context to reduce input cost and latency. The post demonstrates implicit and explicit caching through the Chat Completions API and applies it to an OpenCode coding agent session."
---

# Use prompt caching with open\-weight models on Amazon Bedrock

> [Open the canonical story](<https://aidr.today/7f38ef45?lang=en>)

**Published:** 2026-11-16T15:42:41.000Z
**Category:** Tools
**Topics:** amazon, inference, devtools

## Summary

Prompt caching allows open\-weight models on Amazon Bedrock to reuse long context to reduce input cost and latency\. The post demonstrates implicit and explicit caching through the Chat Completions API and applies it to an OpenCode coding agent session\.

## Sources

- [Story source](<https://aws.amazon.com/blogs/machine-learning/use-prompt-caching-with-open-weight-models-on-amazon-bedrock/>)

