AWS bedrock documentation change

Service: bedrock · 2025-05-13 · Documentation low

File: bedrock/latest/userguide/invoke-imported-model.md

Summary

Added documentation about Converse API limitations for specific models and provided code examples for multi-modal input handling with image placeholders

Security assessment

Changes focus on API usage instructions and model-specific limitations without addressing security vulnerabilities or introducing security features

Diff

diff --git a/bedrock/latest/userguide/invoke-imported-model.md b/bedrock/latest/userguide/invoke-imported-model.md
index 0f06b18b2..92c7c5626 100644
--- a//bedrock/latest/userguide/invoke-imported-model.md
+++ b//bedrock/latest/userguide/invoke-imported-model.md
@@ -12,0 +13,6 @@ After your imported model is available in Amazon Bedrock, you can use the model
+To interface with your imported model using the messages format, you can call the [Converse](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_Converse.html) or [ConverseStream](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_ConverseStream.html) operations. For more information, see [Using the Converse API](./conversation-inference-call.html).
+
+###### Note
+
+Converse API is not supported for Qwen2.5, Qwen2-VL, and Qwen2.5-VL models.
+
@@ -16,0 +23,53 @@ To invoke your imported model, make sure to use the same inference parameters th
+###### Note
+
+When providing multi modal inputs, you will need to include the appropriate placeholders for multi modal tokens in your text prompt. For example, when sending an image input to a Qwen-VL model, the prompt should include `<|vision_start|><|image_pad|><|vision_end|>`. These notations are specific to the model’s tokenizer and can be applied using the following chat template.
+    
+    
+    from transformers import AutoProcessor, AutoTokenizer
+    
+    if vision_model:
+        processor = AutoProcessor.from_pretrained(model)
+    else:
+        processor = AutoTokenizer.from_pretrained(model)
+    
+    
+    # Create messages
+    messages = [
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "image",
+                    "image": "base64 encoded image",
+                },
+                {
+                    "type": "text",
+                    "text": "Describe this image.",
+                },
+            ],
+        }
+    ]
+    
+    # Apply chat template 
+    prompt = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+    """
+    prompt = '''
+    <|im_start|>system\nYou are a helpful assistant.<|im_end|>\n
+    <|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>
+    Describe this image.<|im_end|>\n<|im_start|>assistant\n'''
+    """
+    
+    response = client.invoke_model(
+                    modelId=model_id,
+                    body=json.dumps({
+                        'prompt': prompt,
+                        'temperature': temperature,
+                        'max_gen_len': max_tokens,
+                        'top_p': top_p,
+                        'images': ["base64 encoded image"]  
+                    }),
+                    accept='application/json',
+                    contentType='application/json'
+                ) 
+        
+