Update app.py

app.py

@@ -1,64 +1,120 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-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("HuggingFaceH4/zephyr-7b-beta")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    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})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.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
-"""
-demo = 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 google.generativeai as genai
+import os
+import asyncio  # Import для асинхронности
 
+# Безопасное получение API ключа
+GEMINI_API_KEY = "AIzaSyBoqoPX-9uzvXyxzse0gRwH8_P9xO6O3Bc"
+if not GEMINI_API_KEY:
+    print("Error: GEMINI_API_KEY environment variable not set.")
+    exit()
+
+genai.configure(api_key=GEMINI_API_KEY)
+
+AVAILABLE_MODELS = ["gemini-1.5-flash", "gemini-1.5-pro", "gemini-2.0-flash-thinking-exp"]
+
+# Инициализация моделей один раз при запуске приложения
+MODELS = {model_name: genai.GenerativeModel(model_name=model_name) for model_name in AVAILABLE_MODELS}
+
+async def respond(message, history, selected_model):
+    model = MODELS.get(selected_model)
+    if not model:
+        yield "Error: Selected model not available.", ""
+        return
+
+    try:
+        chat = model.start_chat(history=history)
+        response_stream = chat.send_message(message, stream=True)
+        full_response = ""
+        for chunk in response_stream:
+            full_response += (chunk.text or "")
+            yield full_response, ""  # Пустая строка для thinking output
+    except Exception as e:
+        yield f"Error during API call: {e}", ""
+
+async def respond_thinking(message, history, selected_model):
+    if "thinking" not in selected_model:
+        yield "Thinking model не выбрана.", ""
+        return
+
+    model = MODELS.get(selected_model)
+    if not model:
+        yield "Error: Selected model not available.", ""
+        return
+
+    yield "", "Думаю..." # Сообщение о начале размышлений
+
+    try:
+        response = model.generate_content(message)
+        thinking_process_text = ""
+        model_response_text = ""
+
+        if response.candidates:
+            for part in response.candidates[0].content.parts:
+                if hasattr(part, 'thought') and part.thought == True:
+                    thinking_process_text += f"Model Thought:\n{part.text}\n\n"
+                else:
+                    model_response_text += (part.text or "")
+
+        yield model_response_text, thinking_process_text
+    except Exception as e:
+        yield f"Error during API call: {e}", f"Error during API call: {e}"
+
+def update_chatbot_function(model_name):
+    if "thinking" in model_name:
+        return respond_thinking
+    else:
+        return respond
+
+with gr.Blocks() as demo:
+    gr.Markdown("# Gemini Chatbot с режимом размышления")
+
+    with gr.Row():
+        model_selection = gr.Dropdown(
+            AVAILABLE_MODELS, value="gemini-1.5-flash", label="Выберите модель Gemini"
+        )
+
+    chatbot = gr.ChatInterface(
+        respond,  # Изначально используем асинхронную функцию respond
+        additional_inputs=[model_selection],
+        title="Gemini Chat",
+        description="Общайтесь с моделями Gemini от Google.",
+    )
+
+    thinking_output = gr.Code(label="Процесс размышления (для моделей с размышлением)")
+
+    def change_function(model_name):
+        return update_chatbot_function(model_name)
+
+    model_selection.change(
+        change_function,
+        inputs=[model_selection],
+        outputs=[chatbot],
+    )
+
+    async def process_message(message, history, model_name):
+        if "thinking" in model_name:
+            generator = respond_thinking(message, history, model_name)
+            response, thinking = await generator.__anext__() # Получаем первое значение (пустое сообщение и "Думаю...")
+            yield response, thinking
+            final_response, final_thinking = await generator.__anext__() # Получаем окончательный ответ и размышления
+            yield final_response, final_thinking
+        else:
+            async for response, _ in respond(message, history, model_name):
+                yield response, ""
+
+    chatbot.input_messages[-1].submit(
+        process_message,
+        inputs=[chatbot.input_messages[-1], chatbot.chat_memory, model_selection],
+        outputs=[chatbot.output_messages[-1], thinking_output],
+        scroll_to_output=True,
+    )
+
+    chatbot.input_messages[-1].change(
+        lambda: "",
+        inputs=[],
+        outputs=[thinking_output]
+    )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()