Update app.py

app.py

@@ -1,92 +1,43 @@
+import torch
 import gradio as gr
-from huggingface_hub import InferenceClient
-import os
-
-# Load Hugging Face API Token from Environment Variables
-HF_TOKEN = os.getenv("HF_TOKEN")
-
-# Initialize Mistral-7B Model
-client = InferenceClient("mistralai/Mistral-7B-Instruct-v0.1", token=HF_TOKEN)
-
-# Function to handle chat
-def respond(message, history, system_message, max_tokens, temperature, top_p):
-    messages = [{"role": "system", "content": system_message}]
-
-    for user_msg, bot_msg in history:
-        if user_msg:
-            messages.append({"role": "user", "content": user_msg})
-        if bot_msg:
-            messages.append({"role": "assistant", "content": bot_msg})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=False,  # Change to True if streaming works
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        response += message.choices[0].delta.content
-        yield response
-
-# Custom Styling for Dark Mode
-custom_css = """
-body {
-    background-color: #121212;
-    color: white;
-    font-family: 'Arial', sans-serif;
-}
-
-.gradio-container {
-    max-width: 700px;
-    margin: auto;
-    padding: 20px;
-    background: #1E1E1E;
-    border-radius: 10px;
-    box-shadow: 0px 4px 10px rgba(0, 0, 0, 0.2);
-}
-
-h1 {
-    font-size: 24px;
-    font-weight: bold;
-    text-align: left;
-    color: #00ccff;
-}
-
-h2 {
-    text-align: center;
-    font-size: 30px;
-    font-weight: bold;
-    color: white;
-}
-
-.watermark {
-    text-align: center;
-    font-size: 14px;
-    color: gray;
-    margin-top: 20px;
-}
-"""
-
-# Gradio Chat Interface
-with gr.Blocks(css=custom_css) as demo:
-    gr.Markdown("<h1>Mistral AI Chatbot</h1>")  # Top left title
-    gr.Markdown("<h2>How can I help you?</h2>")  # Center title
-
-    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)"),
-        ],
-    )
-
-    gr.Markdown('<div class="watermark">Created by Rajma</div>')
-
-if __name__ == "__main__":
-    demo.launch()
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+
+model_name = "mistralai/Mistral-7B-Instruct-v0.1"
+
+# Load tokenizer
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+tokenizer.padding_side = "left"
+tokenizer.pad_token = tokenizer.eos_token
+
+# Load model in 4-bit quantization
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    device_map="auto",
+    load_in_4bit=True,
+    torch_dtype=torch.float16
+)
+
+# Create text generation pipeline
+chatbot = pipeline(
+    "text-generation", 
+    model=model, 
+    tokenizer=tokenizer, 
+    device=0, 
+    pad_token_id=tokenizer.eos_token_id
+)
+
+# Function for chatting
+def mistral_chat(user_input):
+    response = chatbot(user_input, max_new_tokens=200, temperature=0.7, do_sample=True)
+    return response[0]["generated_text"]
+
+# Gradio interface
+iface = gr.Interface(
+    fn=mistral_chat,
+    inputs="text",
+    outputs="text",
+    title="Mistral 7B Chatbot"
+)
+
+iface.launch()