Meta’s Llama 3 models are now available in Amazon Bedrock

Today, we are announcing the general availability of Meta’s Llama 3 models in Amazon Bedrock. Meta Llama 3 is designed for you to build, experiment, and responsibly scale your generative artificial intelligence (AI) applications. New Llama 3 models are the most capable to support a broad range of use cases with improvements in reasoning, code generation, and instruction.

According to Meta’s Llama 3 announcement, the Llama 3 model family is a collection of pre-trained and instruction-tuned large language models (LLMs) in 8B and 70B parameter sizes. These models have been trained on over 15 trillion tokens of data—a training dataset seven times larger than that used for Llama 2 models, including four times more code, which supports an 8K context length that doubles the capacity of Llama 2.

You can now use two new Llama 3 models in Amazon Bedrock, further increasing model choice within Amazon Bedrock. These models provide the ability for you to easily experiment with and evaluate even more top foundation models (FMs) for your use case:

• Llama 3 8B is ideal for limited computational power and resources, and edge devices. The model excels at text summarization, text classification, sentiment analysis, and language translation.
• Llama 3 70B is ideal for content creation, conversational AI, language understanding, research development, and enterprise applications. The model excels at text summarization and accuracy, text classification and nuance, sentiment analysis and nuance reasoning, language modeling, dialogue systems, code generation, and following instructions.

Meta is also currently training additional Llama 3 models over 400B parameters in size. These 400B models will have new capabilities, including multimodality, multiple languages support, and a much longer context window. When released, these models will be ideal for content creation, conversational AI, language understanding, research and development (R&D), and enterprise applications.

Llama 3 models in action
If you are new to using Meta models, go to the Amazon Bedrock console and choose Model access on the bottom left pane. To access the latest Llama 3 models from Meta, request access separately for Llama 3 8B Instruct or Llama 3 70B Instruct.

To test the Meta Llama 3 models in the Amazon Bedrock console, choose Text or Chat under Playgrounds in the left menu pane. Then choose Select model and select Meta as the category and Llama 8B Instruct or Llama 3 70B Instruct as the model.

By choosing View API request, you can also access the model using code examples in the AWS Command Line Interface (AWS CLI) and AWS SDKs. You can use model IDs such as meta.llama3-8b-instruct-v1 or meta.llama3-70b-instruct-v1.

Here is a sample of the AWS CLI command:

$ aws bedrock-runtime invoke-model 
  --model-id meta.llama3-8b-instruct-v1:0 
  --body ">system<" 
  --cli-binary-format raw-in-base64-out 
  --region us-east-1 
  invoke-model-output.txt

You can use code examples for Amazon Bedrock using AWS SDKs to build your applications with various programming languages. The following Python code examples show how to invoke the Llama 3 Chat model in Amazon Bedrock for text generation.

def invoke_llama3(self, prompt):
        try:
            body = 
                "prompt": prompt,
                "temperature": 0.5,
                "top_p": 0.9,
                "max_gen_len": 512,
            

            response = self.bedrock_runtime_client.invoke_model(
                modelId="meta.llama3-8b-instruct-v1:0", body=json.dumps(body)
            )

            response_body = json.loads(response["body"].read())
            completion = response_body["generation"]

            return completion

        except ClientError:
            logger.error("Couldn't invoke Llama 3")
            raise

You can use these Llama 3 models for a variety of tasks, such as question answering, language translation, sentiment analysis, and more. Here is an example of Llama 3 models with few-shot in-context learning where we provide training samples available to the model.

Input: Translate English to French:
sea otter => loutre de mer
peppermint => menthe poivrée
plush girafe => girafe peluche
cheese =>
Output:  fromage
panda => panda
penguin => pingouin
polar bear => ours polaire
puffin => macareux
sloth => paresseux
walrus => morse
whale => baleine
zebra => zu00e8bre

Translate English to German:
sea otter => Seeotter

You can also use Llama 3 instruct models optimized for dialogue use cases. The input to the instruct model endpoints is the previous history between the chat assistant and the user. So, you can ask questions contextual to the conversation that has happened so far and provide the system configuration, such as personas, which define the chat assistant’s behavior in the following format:

<


  View Original Article
  

	

	    
		Amazon AI Data & Insights Dept. Dairy Results & Investment Technology & Innovation Predictions and Forecasts Meta (Facebook)	    
	


    

        

            

                
					                
            
            

                
                    
                        AWS Blog
                    
                
				                    
                        Member since
                        
                            October 27, 2023                        
                    
				            

			
	    

        
            
                
	            Loading            
            
                + Follow            
            
                Following