model changed to Haiku 3.5 for query re=write

utilities/invoke_models.py

@@ -59,13 +59,16 @@ def invoke_model_mm(text,img):
     return response_body.get("embedding")
 
 def invoke_llm_model(input,is_stream,model_id="us.anthropic.claude-3-5-haiku-20241022-v1:0"):
-   
+    if(model_id == "us.anthropic.claude-3-5-haiku-20241022-v1:0"):
+        performanceConfigLatency = 'optimized'
+    else:
+        performanceConfigLatency = 'standard'
     if(is_stream == False):
         response = bedrock_runtime_client.invoke_model( 
             modelId= model_id,#"us.anthropic.claude-3-5-haiku-20241022-v1:0",#"anthropic.claude-3-5-haiku-20241022-v1:0",#"anthropic.claude-3-5-sonnet-20240620-v1:0",,
             contentType = "application/json",
             accept = "application/json",
-            performanceConfigLatency='optimized',
+            performanceConfigLatency=performanceConfigLatency,
    
             body = json.dumps({
                         "anthropic_version": "bedrock-2023-05-31",
@@ -128,12 +131,16 @@ def read_from_table(file,question):
     return agent_res
     
 def generate_image_captions_llm(base64_string,question,model_id="us.anthropic.claude-3-5-haiku-20241022-v1:0"):
-    
+    if(model_id == "us.anthropic.claude-3-5-haiku-20241022-v1:0"):
+        performanceConfigLatency = 'optimized'
+    else:
+        performanceConfigLatency = 'standard'
    
     response = bedrock_runtime_client.invoke_model( 
             modelId= model_id,
             contentType = "application/json",
             accept = "application/json",
+            performanceConfigLatency=performanceConfigLatency,
    
             body = json.dumps({
                         "anthropic_version": "bedrock-2023-05-31",