Upload app.py with huggingface_hub

app.py

@@ -1,69 +1,189 @@
+#basic code to play around with the api:
+
+from openai import OpenAI
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
-
-    messages = [{"role": "system", "content": system_message}]
-
-    messages.extend(history)
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from openai import OpenAI
+
+# ==========================================
+# 1. تحميل نموذج Qwen3 محلياً داخل السبيس
+# ==========================================
+print("⏳ loading Qwen .....")
+model_name = "Qwen/Qwen3-0.6B"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto",
+    device_map="auto"
 )
 
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
+# تحديد الشخصية الثابتة للبوت
+SYSTEM_PROMPT = """You are a smart, friendly assistant who answers concisely and accurately.
+    You can understand questions written in either Arabic or English,
+    but you must ALWAYS reply in clear English regardless of the language of the question."""
+
+# ==========================================
+# 2. الدالة الأساسية لمعالجة المحادثة
+# ==========================================
+def chat_engine(user_input, chat_history, model_choice, enable_thinking, api_key):
+    if not user_input.strip():
+        return chat_history, ""
+
+    # تهيئة الذاكرة التراكمية بصيغة الرسائل (Messages Format)
+    messages = [{"role": "system", "content": SYSTEM_PROMPT}]
+
+    # تنظيف التاريخ القديم من تاغات الـ HTML قبل إرساله للنموذج لكي لا تختلط عليه الذاكرة
+    for user_msg, bot_msg in chat_history:
+        messages.append({"role": "user", "content": user_msg})
+        clean_bot_msg = bot_msg.split("</details>")[-1].strip() if "</details>" in bot_msg else bot_msg
+        messages.append({"role": "assistant", "content": clean_bot_msg})
+
+    # إضافة السؤال الحالي
+    messages.append({"role": "user", "content": user_input})
+
+    # ------------------------------------------
+    # المسار الأول: تشغيل نموذج Qwen3 المحلي
+    # ------------------------------------------
+    if model_choice == "Qwen/Qwen3-0.6B":
+        try:
+            # تجهيز التمبلت الخاص بالمحادثة والتفكير
+            text = tokenizer.apply_chat_template(
+                messages,
+                tokenize=False,
+                add_generation_prompt=True,
+                enable_thinking=enable_thinking
+            )
+            model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+
+            # التوليد بحد آمن يناسب السبيس
+            generated_ids = model.generate(
+                **model_inputs,
+                max_new_tokens=1024
+            )
+            output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
+
+            if enable_thinking:
+                try:
+                    # البحث عن تاغ إغلاق التفكير 151668 (</think>)
+                    index = len(output_ids) - output_ids[::-1].index(151668)
+                except ValueError:
+                    index = 0
+
+                thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
+                assistant_content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
+
+                # صياغة النص بـ Markdown لإظهار التفكير بشكل منسدل أنيق
+                if thinking_content:
+                    final_output = f"<details><summary>🧠 كواليس تفكير النموذج (اضغط للتوسيع)</summary>\n\n{thinking_content}\n\n</details>\n\n{assistant_content}"
+                else:
+                    final_output = assistant_content
+            else:
+                final_output = tokenizer.decode(output_ids, skip_special_tokens=True).strip("\n")
+        except Exception as e:
+            final_output = f"❌ خطأ أثناء توليد رد النموذج المحلي:\n{str(e)}"
+
+        chat_history.append((user_input, final_output))
+        return chat_history, ""
+
+    # ------------------------------------------
+    # المسار الثاني: تشغيل GPT-5 عبر الـ API الخارجي
+    # ------------------------------------------
+    elif model_choice == "GPT-5 (OpenAI API)":
+        if not api_key.strip():
+            chat_history.append((user_input, "⚠️ الرجاء إدخال مفتاح OpenAI API في خانة الإعدادات لاستخدام GPT-5."))
+            return chat_history, ""
+
+        try:
+            # إنشاء اتصال مع OpenAI باستخدام الـ API الممرر من الواجهة
+            client = OpenAI(api_key=api_key)
+
+            response = client.chat.completions.create(
+                model="gpt-5", # اسم الموديل الخاص بالـ API لديك
+                messages=messages,
+                max_completion_tokens=4096
+            )
 
+            gpt_response = response.choices[0].message.content
+            chat_history.append((user_input, gpt_response))
+        except Exception as e:
+            chat_history.append((user_input, f"❌ connecation to API faild :\n{str(e)}"))
 
+        return chat_history, ""
+
+# ==========================================
+# 3. (Gradio UI Layout)
+# ==========================================
+with gr.Blocks(theme=gr.themes.Soft(primary_hue="blue", secondary_hue="slate")) as demo:
+    gr.Markdown("# 🤖 Multi Model chat bot (Level 4)")
+    gr.Markdown("changing model and thinking mode ")
+
+    with gr.Row():
+        # القائمة الجانبية للتحكم والإعدادات
+        with gr.Column(scale=1, min_width=280):
+            gr.Markdown("### ⚙️ sting")
+
+            model_dropdown = gr.Dropdown(
+                choices=["Qwen/Qwen3-0.6B", "GPT-5 (OpenAI API)"],
+                value="Qwen/Qwen3-0.6B",
+                label="choose active model"
+            )
+
+            thinking_checkbox = gr.Checkbox(
+                label="activate thinking mode (for Qwen)",
+                value=True
+            )
+
+            api_key_input = gr.Textbox(
+                label="OpenAI API Key (لـ GPT-5)",
+                placeholder="sk-...",
+                type="password"
+            )
+
+            clear_button = gr.Button("🧹 clear chat", variant="secondary")
+
+        # الشاشة الرئيسية للمحادثة
+        with gr.Column(scale=4):
+            chatbot = gr.Chatbot(
+                label="CHAT",
+                height=550,
+                show_copy_button=True
+            )
+
+            with gr.Row():
+                user_msg = gr.Textbox(
+                    label="ask your question and press ENTER",
+                    placeholder="What do yiu want for christmas?",
+                    scale=4
+                )
+                submit_button = gr.Button("send🚀", variant="primary", scale=1)
+
+    # ------------------------------------------
+    # ربط العناصر والأحداث (Event Handling)
+    # ------------------------------------------
+
+    # التفاعل عند الضغط على زر إرسال أو Enter
+    submit_event = submit_button.click(
+        chat_engine,
+        inputs=[user_msg, chatbot, model_dropdown, thinking_checkbox, api_key_input],
+        outputs=[chatbot, user_msg]
+    )
+
+    user_msg.submit(
+        chat_engine,
+        inputs=[user_msg, chatbot, model_dropdown, thinking_checkbox, api_key_input],
+        outputs=[chatbot, user_msg]
+    )
+
+    # تفاعل زر مسح الذاكرة
+    clear_button.click(
+        fn=lambda: [],
+        inputs=None,
+        outputs=chatbot,
+        queue=False
+    )
+
+# تشغيل التطبيق
 if __name__ == "__main__":
     demo.launch()
+