formatted

app.py

@@ -7,25 +7,24 @@ For more information on `huggingface_hub` Inference API support, please check th
 """
 client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1",token=os.getenv('HUGGINGFACE_TOKEN').strip())
 
+def format_prompt(message, history):
+  prompt = "<s>"
+  for user_prompt, bot_response in history:
+    prompt += f"[INST] {user_prompt} [/INST]"
+    prompt += f" {bot_response}</s> "
+  prompt += f"[INST] {message} [/INST]"
+  return prompt
+
 def generate_response(
-    message,
+    prompt,
     history: list[tuple[str, str]],
-    system_message,
+    system_prompt,
     max_tokens,
     temperature,
     top_p,
 ):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    stream = client.text_generation(message,stream=True, max_new_tokens=256)
+    formatted_prompt = format_prompt(f"{system_prompt}, {prompt}", history)
+    stream = client.text_generation(formatted_prompt,stream=True, max_new_tokens=256)
     output = ""
     for response in stream:
         output += response