Update app.py

app.py

@@ -19,6 +19,9 @@ model = AutoModelForCausalLM.from_pretrained(model_id)
 # Create a text generation pipeline
 pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
 
+# Define the pre-prompt
+PRE_PROMPT = "You are a helpful virtual assistant. Answer the user's question clearly and concisely."
+
 # Define request body schema
 class TextGenerationRequest(BaseModel):
     prompt: str
@@ -33,13 +36,19 @@ class TextGenerationRequest(BaseModel):
 async def generate_text(request: TextGenerationRequest):
     try:
         logger.info("Generating text...")
+        
+        # Combine the pre-prompt and user's prompt
+        combined_input = f"{PRE_PROMPT} {request.prompt}"
+        
+        # Generate text using the pipeline
         outputs = pipe(
-            request.prompt,
+            combined_input,  # Use the combined input
             max_new_tokens=request.max_new_tokens,
             temperature=request.temperature,
             top_k=request.top_k,
             top_p=request.top_p,
-            do_sample=request.do_sample
+            do_sample=request.do_sample,
+            return_full_text=False  # Exclude the input prompt from the output
         )
         return {"generated_text": outputs[0]["generated_text"]}
     except Exception as e: