Update app.py

app.py

@@ -2,13 +2,9 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 import os
 
-
-
 API_TOKEN = os.getenv("HF_TOKEN")
-
 client = InferenceClient(token=API_TOKEN)
 
-
 def respond(
     message,
     history: list[tuple[str, str]],
@@ -17,31 +13,31 @@ def respond(
     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})
+    # بناء نص المحادثة ك string واحد مع أدوار واضحة
+    conversation = f"System: {system_message}\n"
+    for user_msg, assistant_msg in history:
+        if user_msg:
+            conversation += f"User: {user_msg}\n"
+        if assistant_msg:
+            conversation += f"Assistant: {assistant_msg}\n"
+    conversation += f"User: {message}\nAssistant:"
 
     response = ""
 
-    for message in client.chat_completion(
-        model="Alhdrawi/alhdrawi",  
-        messages=messages,
-        max_tokens=max_tokens,
-        stream=True,
+    # استدعاء text_generation مع التدفق (stream=True)
+    for output in client.text_generation(
+        model="Alhdrawi/alhdrawi",
+        inputs=conversation,
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
+        stream=True,
     ):
-        token = message.choices[0].delta.content
-        response += token
+        # كل مرة يجي جزء جديد من النص
+        new_text = output.generated_text[len(response):]
+        response += new_text
         yield response
 
-
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[