AWS bedrock documentation change

Service: bedrock · 2026-04-22 · Documentation low

File: bedrock/latest/userguide/model-card-amazon-nova-2-lite.md

Summary

Added prompt caching documentation with technical specifications and updated API example structure

Security assessment

This change adds performance optimization documentation (prompt caching) and updates API examples to use new message format with inferenceConfig. No security-related content is added or modified; this is purely feature documentation and example updates.

Diff

diff --git a/bedrock/latest/userguide/model-card-amazon-nova-2-lite.md b/bedrock/latest/userguide/model-card-amazon-nova-2-lite.md
index 8ddc885b4..68e44946c 100644
--- a//bedrock/latest/userguide/model-card-amazon-nova-2-lite.md
+++ b//bedrock/latest/userguide/model-card-amazon-nova-2-lite.md
@@ -67,0 +68,10 @@ Nova 2 Lite is Amazon's cost-efficient multimodal model for simple automation, d
+**Prompt caching using`bedrock-runtime` endpoint**
+
+For more information, see [Prompt caching for faster model inference](https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-caching.html).
+
+**Prompt caching supported** | **Min tokens per cache checkpoint** | **Max cache checkpoints per request** | **Supported TTL** | **Fields that accept prompt cache checkpoints**  
+---|---|---|---|---  
+Yes | 1K* | 4 | 5 minutes | `system` and `messages`  
+  
+_* Amazon Nova models support a maximum of 20K tokens for prompt caching. Prompt caching is primarily for text prompts._
+
@@ -227,2 +237,7 @@ Invoke API
-                'messages': [{ 'role': 'user', 'content': 'Can you explain the features of Amazon Bedrock?'}],
-                'max_tokens': 1024
+                'messages': [{
+                    'role': 'user',
+                    'content': [{'text': 'Can you explain the features of Amazon Bedrock?'}]
+                }],
+                'inferenceConfig': {
+                    'maxTokens': 1024
+                }