Update app.py

app.py

@@ -52,23 +52,20 @@ def format_prompt(message, history):
 
 def generate(prompt, history=[], temperature=0.1, max_new_tokens=10000, top_p=0.95, repetition_penalty=1.0):
     global total_tokens_used
-    input_tokens = len(tokenizer.encode(prompt))
-    total_tokens_used += input_tokens
-    available_tokens = 32768 - total_tokens_used
-    if available_tokens <= 0:
-        return f"Error: 입력이 최대 허용 토큰 수를 초과합니다. Total tokens used: {total_tokens_used}"
-
-    formatted_prompt = format_prompt(prompt, history)
-    output_accumulated = ""
+    input_tokens = tokenizer.encode(prompt)
+    total_tokens_used += len(input_tokens)
+    if total_tokens_used >= 32768:
+        return "Error: 입력이 최대 허용 토큰 수를 초과하였습니다."
     try:
-        stream = client.text_generation(formatted_prompt, temperature=temperature, max_new_tokens=min(max_new_tokens, available_tokens),
-                                        top_p=top_p, repetition_penalty=repetition_penalty, do_sample=True, seed=42, stream=True)
-        for response in stream:
-            output_part = response['generated_text'] if 'generated_text' in response else str(response)
-            output_accumulated += output_part
-        return output_accumulated + f"\n\n---\nTotal tokens used: {total_tokens_used}"
+        response = client(text=prompt, temperature=temperature, max_tokens=max_new_tokens)
+        response_text = response.get('generated_text', '')
+        if "티커" in prompt:
+            ticker = prompt.split()[-1]
+            response_text += "\n" + fetch_ticker_info(ticker)
+        return response_text
     except Exception as e:
-        return f"Error: {str(e)}\nTotal tokens used: {total_tokens_used}"
+        return f"오류 발생: {str(e)}"
+
 
 def postprocess(history):
     user_prompt = history[-1][0]