Microsoft model

app.py

@@ -2,8 +2,8 @@ import gradio as gr
 from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
 
 # Load the model and tokenizer using Hugging Face
-#model_name = "microsoft/Phi-3-mini-4k-instruct"
-model_name = "KingNish/Qwen2.5-0.5b-Test-ft"
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+#model_name = "KingNish/Qwen2.5-0.5b-Test-ft"
 
 
 # Explicitly load the tokenizer and model
@@ -23,8 +23,9 @@ def respond(
     top_p,
 ):
     # Combine system message and conversation history
-    prompt = system_message + "\n"
-    prompt += f"User: {message}\n\nBot:"
+    prompt=message
+    #prompt = system_message + "\n"
+    #prompt += f"User: {message}\n\nBot:"
 
     # Generate the response using the model
     response = chatbot(prompt, max_length=max_tokens, temperature=temperature, top_p=top_p)[0]['generated_text']