nova model added

Author: jfbaeta

README.md

@@ -90,14 +90,16 @@ For a more accurate cost estimate, use the AWS Pricing Calculator and input your
 
 ### Supported AI Models
 
-The system supports three different AI models for chat moderation:
+The system supports four different AI models for chat moderation:
 
 1. **Anthropic Claude Haiku**: Claude 3 Haiku is Anthropic's fastest, most compact model for near-instant responsiveness. It answers simple queries and requests with speed. Customers will be able to build seamless AI experiences that mimic human interactions. Claude 3 Haiku can process images and return text outputs, and features a 200K context window.
 
 2. **Amazon Titan**: Amazon Titan Text Premier is an advanced, high-performance, and cost-effective LLM engineered to deliver superior performance for enterprise-grade text generation applications, including optimized performance for retrieval-augmented generation (RAG) and Agents.
 
 3. **Meta Llama**: Meta Llama 3 is an accessible, open large language model (LLM) designed for developers, researchers, and businesses to build, experiment, and responsibly scale their generative AI ideas. Part of a foundational system, it serves as a bedrock for innovation in the global community. Ideal for limited computational power and resources, edge devices, and faster training times.
 
+4. **Amazon Nova Micro**: Amazon Nova family of models offer customers multiple price performance operating points to best optimize between accuracy, speed, and cost. Amazon Nova Micro is a text only model that delivers the lowest latency responses at the lowest cost per inference among Nova family.
+
 Each model has its own strengths and characteristics. You can switch between these models using the `prompt-switch.bash` script.
 
 ## Prerequisites
@@ -222,7 +224,7 @@ To switch between different AI models or prompts:
 ./prompt-switch.bash <model-name>
 ```
 
-Replace `<model-name>` with one of the available options: `titan`, `haiku`, or `llama`. The aforementioned `./install.bash` script configures Anthropic Claude Haiku to be used by default.
+Replace `<model-name>` with one of the available options: `titan`, `haiku`, `llama` or `nova-micro`. The aforementioned `./install.bash` script configures Anthropic Claude Haiku to be used by default.
 
 ### Updating the Front-End
 
@@ -234,6 +236,10 @@ After making changes to the front-end code, deploy updates using:
 
 This script will build the React application and update the S3 bucket and CloudFront distribution. This is not required at the first deployment.
 
+### Scripts Usage
+
+The scripts should be executed from the `./scripts` directory. Running from another path may cause issues or the scripts may fail.
+
 ### Moderation Guidelines
 
 The AI models use the following prompt as the main guideline for moderating chat messages:

backend/cdk/lib/api-stack.ts

@@ -306,6 +306,7 @@ export class Api extends cdk.NestedStack {
                 `arn:aws:bedrock:${this.region}:*:foundation-model/amazon.titan-text-premier-v1:0`,
                 `arn:aws:bedrock:${this.region}:*:foundation-model/anthropic.claude-3-haiku-20240307-v1:0`,
                 `arn:aws:bedrock:${this.region}:*:foundation-model/meta.llama3-8b-instruct-v1:0`,
+                `arn:aws:bedrock:${this.region}:*:foundation-model/amazon.nova-micro-v1:0`
               ],
             }),
           ],

backend/cdk/lib/main-stack.ts

@@ -65,6 +65,7 @@ export class MainStack extends cdk.Stack {
       "amazon.titan-text-premier-v1:0",
       "anthropic.claude-3-haiku-20240307-v1:0",
       "meta.llama3-8b-instruct-v1:0",
+      "amazon.nova-micro-v1:0"
     ];
 
     const observability = new Observability(this, "Observability", {

scripts/insert-prompt.bash

@@ -2,7 +2,7 @@
 
 # Check if an argument is provided
 if [ -z "$1" ]; then
-  echo -e "\n${RED}[ERROR] Please provide a model name (titan, haiku, or llama) as an argument."
+  echo -e "\n${RED}[ERROR] Please provide a model name (titan, haiku, llama or nova-micro) as an argument."
   exit 1
 fi
 
@@ -51,8 +51,16 @@ case "$1" in
     TEMPERATURE=0
     TOP_P=0
     ;;
+  nova-micro)
+    MODEL_ID="amazon.nova-micro-v1:0"
+    MODEL_NAME="Amazon Nova Micro"
+    MODEL_OUTPUT_KEY="NovaMicroModelUUID"
+    MAX_TOKENS=256
+    TEMPERATURE=0
+    TOP_P=0
+    ;;
   *)
-    echo -e "\n${RED}[ERROR] Invalid model name provided. Please use titan, haiku, or llama."
+    echo -e "\n${RED}[ERROR] Invalid model name provided. Please use titan, haiku, llama or nova-micro."
     exit 1
     ;;
 esac
@@ -102,4 +110,4 @@ if [ $? -eq 0 ]; then
 else
   echo -e "\n${RED}[ERROR] Failed to insert ${MODEL_NAME} model prompt.${NC}"
   exit 1
-fi
+fi

scripts/install.bash

@@ -147,7 +147,9 @@ update_and_run_scripts() {
     run_script ./insert-prompt.bash haiku
 
     run_script ./insert-prompt.bash llama
-    
+
+    run_script ./insert-prompt.bash nova-micro
+
     run_script ./prompt-switch.bash haiku
 
     run_script ./publish.bash

scripts/prompt-switch.bash

@@ -10,6 +10,7 @@ STACK_NAME=$(jq -r 'keys[0]' "$CDK_OUTPUTS_FILE")
 TITAN_MODEL_UUID=$(jq -r '.'"$STACK_NAME"'.TitanModelUUID' "$CDK_OUTPUTS_FILE")
 HAIKU_MODEL_UUID=$(jq -r '.'"$STACK_NAME"'.HaikuModelUUID' "$CDK_OUTPUTS_FILE")
 LLAMA_MODEL_UUID=$(jq -r '.'"$STACK_NAME"'.LlamaModelUUID' "$CDK_OUTPUTS_FILE")
+NOVA_MICRO_MODEL_UUID=$(jq -r '.'"$STACK_NAME"'.NovaMicroModelUUID' "$CDK_OUTPUTS_FILE")
 PROMPT_SWITCH_PARAMETER_NAME=$(jq -r '.'"$STACK_NAME"'.PromptSwitchParameterName' "$CDK_OUTPUTS_FILE")
 
 # Color variables
@@ -36,8 +37,11 @@ case "$1" in
   llama)
     NEW_ACTIVE_MODEL="$LLAMA_MODEL_UUID"
     ;;
+  nova-micro)
+    NEW_ACTIVE_MODEL="$NOVA_MICRO_MODEL_UUID"
+    ;;
   *)
-    echo -e "\n${RED}[ERROR] Invalid model name provided. Please use titan, haiku, or llama."
+    echo -e "\n${RED}[ERROR] Invalid model name provided. Please use titan, haiku, llama or nova-micro."
     exit 1
     ;;
 esac
@@ -57,4 +61,4 @@ if [ $? -ne 0 ]; then
 fi
 
 echo -e "\n${YELLOW}[WARNING] Model and prompt switched to: $1 (UUID: $NEW_ACTIVE_MODEL)"
-echo -e "\n${GREEN}[SUCCESS] Model and prompt switch completed successfully!${NC}"
+echo -e "\n${GREEN}[SUCCESS] Model and prompt switch completed successfully!${NC}"

scripts/publish.bash

@@ -88,4 +88,4 @@ aws cloudfront wait invalidation-completed --distribution-id "$CLOUDFRONT_DISTRI
 check_status "CloudFront invalidation completion"
 
 echo -e "\n${GREEN}[SUCCESS] Front-End Environment published successfully.${NC}"
-echo -e "\n${BLUE}[INFO] CloudFront Distribution Domain: https://${CLOUDFRONT_DISTRIBUTION_DOMAIN}${NC}"
+echo -e "\n${BLUE}[INFO] CloudFront Distribution Domain: https://${CLOUDFRONT_DISTRIBUTION_DOMAIN}${NC}"