AWS bedrock documentation change

Service: bedrock · 2025-05-03 · Documentation low

File: bedrock/latest/userguide/distillation-prepare-datasets.md

Summary

Restructured content about supported distillation modalities, removed model support table, added detailed examples for RAG/agent prompt formatting, and updated section headers

Security assessment

Changes focus on improving documentation structure and providing implementation examples. No security vulnerabilities or security features are mentioned. The prompt examples demonstrate functional use cases without security implications.

Diff

diff --git a/bedrock/latest/userguide/distillation-prepare-datasets.md b/bedrock/latest/userguide/distillation-prepare-datasets.md
index 5ec8911bb..ea5b50d22 100644
--- a//bedrock/latest/userguide/distillation-prepare-datasets.md
+++ b//bedrock/latest/userguide/distillation-prepare-datasets.md
@@ -5 +5 @@
-Supported modalities for distillation
+Supported modalities for distillationOptimize your input prompts for synthetic data generation
@@ -17,20 +16,0 @@ Amazon Nova models have different requirements for distillation. For more inform
-## Supported modalities for distillation
-
-The following table shows the input and output modalities that distillation supports with each model:
-
-Model name | Text-to-text  
----|---  
-Amazon Nova Pro | Yes  
-Amazon Nova Lite | Yes  
-Amazon Nova Micro | Yes  
-Amazon Titan Text G1 - Express | No  
-Amazon Titan Text G1 - Lite | No  
-Amazon Titan Text Premier | No  
-Amazon Titan Image Generator G1 V1 | No  
-Amazon Titan Multimodal Embeddings G1 G1 | No  
-Anthropic Claude 3 Haiku | Yes  
-Cohere Command | No  
-Cohere Command Light | No  
-Meta Llama 2 13B | No  
-Meta Llama 2 70B | No  
-  
@@ -38,0 +19 @@ Meta Llama 2 70B | No
+  * Supported modalities for distillation
@@ -40 +21 @@ Meta Llama 2 70B | No
-
+  * Optimize your input prompts for synthetic data generation
@@ -48,0 +30,87 @@ Meta Llama 2 70B | No
+## Supported modalities for distillation
+
+The models listed in [Supported models and Regions for Amazon Bedrock Model Distillation](./prequisites-model-distillation.html#model-distillation-supported) support only the text-to-text modality.
+
+## Optimize your input prompts for synthetic data generation
+
+During model distillation, Amazon Bedrock generates a synthetic dataset that it uses to fine tune your student model for your specific use case. For more information, see [How Amazon Bedrock Model Distillation works](./model-distillation.html#how-md-works).
+
+You can optimize the synthetic data generation process by formatting your input prompts for the use case that you want. For example, if your distilled model's use case is retrieval augmented generation (RAG), you would format your prompts differently than if you want the model to focus on agent use cases.
+
+The following are examples for how you can format your input prompts for RAG or agent use cases.
+
+RAG prompt example
+    
+    
+    
+    {
+      "schemaVersion": "bedrock-conversation-2024",
+      "system": [
+        {
+          "text": "You are a financial analyst charged with answering questions about 10K and 10Q SEC filings. Given the context below, answer the following question."
+        }
+      ],
+      "messages": [
+        {
+          "role": "user",
+          "content": [
+            {
+              "text": "<context>\nDocument 1: Multiple legal actions have been filed against us as a result of the October 29, 2018 accident of Lion Air Flight 610 and the March 10, 2019 accident of Ethiopian Airlines Flight 302.\n</context>\n\n<question>Has Boeing reported any materially important ongoing legal battles from FY2022?</question>"
+            }
+          ]
+        }
+      ]
+    }
+
+Agent prompt example
+    
+    
+    
+    {
+        "schemaVersion": "bedrock-conversation-2024",
+        "system": [
+            {
+                "text": 'You are an expert in composing functions. You are given a question and a set of possible functions. Based on the question, you will need to make one or more function/tool calls to achieve the purpose.
+                        Here is a list of functions in JSON format that you can invoke.
+                        [
+                            {
+                                "name": "lookup_weather",
+                                "description: "Lookup weather to a specific location",
+                                "parameters": {
+                                    "type": "dict",
+                                    "required": [
+                                        "city"
+                                    ],
+                                    "properties": {
+                                        "location": {
+                                            "type": "string",
+                                        },
+                                        "date": {
+                                            "type": "string",
+                                        }
+                                    }
+                                }
+                            }
+                        ]'
+            }
+        ],
+        "messages": [
+            {
+                "role": "user",
+                "content": [
+                    {
+                        "text": "What's the weather tomorrow?"
+                    }
+                ]
+            },
+            {
+                "role": "assistant",
+                "content": [
+                   {
+                       "text": "[lookup_weather(location=\"san francisco\", date=\"tomorrow\")]"
+                   }
+                ]
+            }
+        ]
+    }
+
@@ -55 +123 @@ To use the Amazon Web Services Documentation, Javascript must be enabled. Please
-Decide on your teacher and student models for distillation
+Choose teacher and student models for distillation