diff --git a/src/interface.jl b/src/interface.jl
index 1591264..d24b72e 100644
--- a/src/interface.jl
+++ b/src/interface.jl
@@ -198,7 +198,17 @@ function decisionMaker(state::T1, context, text2textInstructLLM::Function,
   header = ["Comprehension:", "Plan:", "Action_name:", "Action_input:"]
   dictkey = ["comprehension", "plan", "action_name", "action_input"]
 
+  llmkwargs=Dict(
+      :num_ctx => 32768,
+      :temperature => 0.1,
+    )
+
   for attempt in 1:10
+    if attempt > 1
+      println("\nERROR SQLLLM decisionMaker() attempt $attempt/10 ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
+      llmkwargs[:temperature] = 0.1 * attempt
+    end
+
     QandA = generatequestion(state, context, text2textInstructLLM; similarSQL=similarSQL_)
 
     usermsg =
@@ -220,8 +230,8 @@ function decisionMaker(state::T1, context, text2textInstructLLM::Function,
     ]
 
     # put in model format
-    prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct")
-    response = text2textInstructLLM(prompt, modelsize="medium")
+    prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen")
+    response = text2textInstructLLM(prompt; llmkwargs=llmkwargs)
 
     # LLM tends to generate observation given that it is in the input
     response = 
@@ -249,8 +259,8 @@ function decisionMaker(state::T1, context, text2textInstructLLM::Function,
     end
 
     if occursin("NULL", response)
-      errornote = "\nSQL decisionMaker() NULL response is not allowed"
-      println("Attempt $attempt $errornote ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
+      errornote = "\nYour previous attempt was NULL. This is not allowed"
+      println("\nERROR SQLLLM decisionMaker() $errornote ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
       continue
     end
 
@@ -272,10 +282,12 @@ function decisionMaker(state::T1, context, text2textInstructLLM::Function,
     # check whether response has all header
     detected_kw = GeneralUtils.detect_keyword(header, response)
     if 0 ∈ values(detected_kw)
-      errornote = "\nSQLLLM decisionMaker() response does not have all header"
+      errornote = "\nYour previous attempt did not have all points according to the required response format"
+      println("\nERROR SQLLLM decisionMaker() $errornote \n$response", @__FILE__, ":", @__LINE__, " $(Dates.now())")
       continue
     elseif sum(values(detected_kw)) > length(header)
-      errornote = "\nSQLLLM decisionMaker() response has duplicated header"
+      errornote = "\nYour previous attempt has duplicated points according to the required response format"
+      println("\nERROR SQLLLM decisionMaker() $errornote \n$response", @__FILE__, ":", @__LINE__, " $(Dates.now())")
       continue
     end
 
@@ -298,43 +310,33 @@ function decisionMaker(state::T1, context, text2textInstructLLM::Function,
 
     toollist = ["TABLEINFO", "RUNSQL"]
     if responsedict[:action_name] ∉ toollist
-      errornote = "\nYou must only use the given functions"
-      println("Attempt $attempt $errornote ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
+      errornote = "\nYour previous attempt has action_name that is not in the tool list"
+      println("\nERROR SQLLLM decisionMaker() $errornote ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
       continue
     end
 
     for i in toollist
       if occursin(i, responsedict[:action_input])
-        errornote = "\n action_name is in action_input which is not allowed."
-        println("Attempt $attempt $errornote ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
+        errornote = "\nYour previous attempt has action_name in action_input which is not allowed"
+        println("\nERROR SQLLLM decisionMaker() $errornote ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
         continue
       end
     end
 
     for i ∈ Symbol.(dictkey)
       if length(JSON3.write(responsedict[i])) == 0
-        errornote = "\n $i is empty"
-        println("Attempt $attempt $errornote ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
+        errornote = "\nYour previous attempt has empty value for $i"
+        println("\nERROR SQLLLM decisionMaker() $errornote ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
         continue
       end
     end
 
-    # check whether response has all header
-    detected_kw = GeneralUtils.detect_keyword(header, response)
-    if 0 ∈ values(detected_kw)
-      errornote = "\nSQL decisionMaker() response does not have all header"
-      continue
-    elseif sum(values(detected_kw)) > length(header)
-      errornote = "\nSQL decisionMaker() response has duplicated header"
-      continue
-    end
-
     state[:decisionMaker] = responsedict
 
     return responsedict
 
   end
-  error("DecisionMaker failed to generate a thought \n", response)
+  error("SQLLLM DecisionMaker() failed to generate a thought \n", response)
 end
 
 
@@ -446,7 +448,7 @@ function evaluator(state::T1, text2textInstructLLM::Function
     ]
 
     # put in model format
-    prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct")
+    prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen")
 
     header = ["Trajectory_evaluation:", "Answer_evaluation:", "Accepted_as_answer:", "Score:", "Suggestion:"]
     dictkey = ["trajectory_evaluation", "answer_evaluation", "accepted_as_answer", "score", "suggestion"]
@@ -601,7 +603,7 @@ function reflector(config::T1, state::T2)::String where {T1<:AbstractDict, T2<:A
   ]
 
   # put in model format
-  prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct")
+  prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen")
   externalService = config[:externalservice][:text2textinstruct]
 
   # apply LLM specific instruct format
@@ -1011,7 +1013,7 @@ function query(query::T, executeSQL::Function, text2textInstructLLM::Function;
     #   JSON3.pretty(io, highValueState)
     # end
     selected = compareState(query, highValueState, text2textInstructLLM)
-    resultState = highValueState[selected]
+    resultState = highValueState[selected]  #BUG compareState() select 0
   end
   latestKey, latestInd = GeneralUtils.findHighestIndexKey(resultState[:thoughtHistory], "observation")
   action_input = Symbol("action_input_$latestInd")  # latest sql
@@ -1029,7 +1031,9 @@ function query(query::T, executeSQL::Function, text2textInstructLLM::Function;
     println("query() return nothing")
   end
 
-  return (text=extracted, rawresponse=resultState[:rawresponse]) 
+  result = (text=extracted, rawresponse=resultState[:rawresponse])
+
+  return result 
 end
 
 
@@ -1172,7 +1176,7 @@ function generatequestion(state::T1, context, text2textInstructLLM::Function;
     ]
 
     # put in model format
-    prompt = GeneralUtils.formatLLMtext(_prompt; formatname="llama3instruct")
+    prompt = GeneralUtils.formatLLMtext(_prompt; formatname="qwen")
 
     try
       response = text2textInstructLLM(prompt, modelsize="medium")
@@ -1190,7 +1194,7 @@ function generatequestion(state::T1, context, text2textInstructLLM::Function;
       responsedict = GeneralUtils.textToDict(response, header; 
                                             dictKey=dictkey, symbolkey=true)
       response = "Q1: " * responsedict[:q1]
-      # println("\nSQLLLM generatequestion() ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
+      # println("\nERROR SQLLLM generatequestion() ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
       pprintln(Dict(responsedict))
       return response
     catch e
@@ -1198,7 +1202,7 @@ function generatequestion(state::T1, context, text2textInstructLLM::Function;
       showerror(io, e)
       errorMsg = String(take!(io))
       st = sprint((io, v) -> show(io, "text/plain", v), stacktrace(catch_backtrace()))
-      println("\nSQLLLM generatequestion() Attempt $attempt. Error occurred: $errorMsg\n$st ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
+      println("\nERROR SQLLLM generatequestion() Attempt $attempt. Error occurred: $errorMsg\n$st ", @__FILE__, ":", @__LINE__, " $(Dates.now())")
     end
   end
   error("generatequestion failed to generate a thought ", response)
diff --git a/system_prompt_template.jl b/system_prompt_template.jl
index a3d1d24..158c943 100644
--- a/system_prompt_template.jl
+++ b/system_prompt_template.jl
@@ -22,9 +22,9 @@ Default system message template:
   <Your responsibility does NOT includes>
   - 
   </Your responsibility does NOT includes>
-  <At each round of conversation, you will be given the following>
+  <At each round of conversation, you will be given the following information>
 
-  </At each round of conversation, you will be given the following>
+  </At each round of conversation, you will be given the following information>
   <You must follow the following guidelines>
     -
   </You must follow the following guidelines>