Update app.py

app.py

@@ -1,64 +1,129 @@
-import gradio as gr
-from huggingface_hub import InferenceClient
+# NORTHERN_AI 
+# Created by AR.BALTEE
 
-"""
-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")
 
+import os
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from flask import Flask, request, jsonify
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+app = Flask(__name__)
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Choose a lightweight open model that can run on limited hardware
+# Options include:
+# - GPT2-small (if you have ~2GB RAM for the model)
+# - Hugging Face's inference endpoints (cloud-based, some free tiers)
+# - Models like DialoGPT-small, BLOOM-560M, or OPT-350M
 
-    messages.append({"role": "user", "content": message})
+# Configuration
+MODEL_NAME = "EleutherAI/gpt-neo-125M"  # A relatively small model, replace with your choice
+USE_CLOUD_INFERENCE = True  # Set to True to use Hugging Face's Inference API instead of local model
 
-    response = ""
+# Hugging Face API Token (sign up for free at huggingface.co)
+HF_API_TOKEN = os.environ.get("HF_API_TOKEN", "")  # Store your token as an environment variable for security
 
-    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
+# System prompt that defines your AI assistant's personality
+SYSTEM_PROMPT = """NORTHERN_AI is a helpful AI assistant created by AR.BALTEE.
+It aims to provide accurate and helpful information to users' questions.
+NORTHERN_AI is friendly, concise, and knowledgeable."""
 
-        response += token
-        yield response
+class NorthernAI:
+    def __init__(self):
+        self.system_prompt = SYSTEM_PROMPT
+        
+        if not USE_CLOUD_INFERENCE:
+            print("Loading model locally (requires sufficient RAM)...")
+            self.tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+            # Load in 8-bit to reduce memory requirements
+            self.model = AutoModelForCausalLM.from_pretrained(
+                MODEL_NAME, 
+                torch_dtype=torch.float16,
+                low_cpu_mem_usage=True,
+                device_map="auto"
+            )
+        else:
+            print("Using cloud inference API (minimal RAM required)...")
+            # For cloud inference, we'll just need the API endpoint
+            from huggingface_hub import InferenceClient
+            self.client = InferenceClient(token=HF_API_TOKEN)
+    
+    def generate_response(self, user_input):
+        prompt = f"{self.system_prompt}\n\nUser: {user_input}\nNORTHERN_AI:"
+        
+        if USE_CLOUD_INFERENCE:
+            # Use Hugging Face's Inference API
+            response = self.client.text_generation(
+                prompt,
+                model=MODEL_NAME,
+                max_new_tokens=150,
+                temperature=0.7,
+                top_p=0.95,
+                repetition_penalty=1.1
+            )
+            return response.strip()
+        else:
+            # Local generation
+            inputs = self.tokenizer(prompt, return_tensors="pt").to(self.model.device)
+            with torch.no_grad():
+                output = self.model.generate(
+                    **inputs,
+                    max_new_tokens=150,
+                    temperature=0.7,
+                    top_p=0.95,
+                    repetition_penalty=1.1
+                )
+            return self.tokenizer.decode(output[0], skip_special_tokens=True).split("NORTHERN_AI:")[-1].strip()
 
+# Initialize the AI assistant
+northern_ai = NorthernAI()
 
-"""
-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)",
-        ),
-    ],
-)
+@app.route('/api/chat', methods=['POST'])
+def chat():
+    data = request.json
+    user_message = data.get('message', '')
+    response = northern_ai.generate_response(user_message)
+    return jsonify({"response": response})
 
+@app.route('/')
+def home():
+    return """
+    <html>
+        <head><title>NORTHERN_AI by AR.BALTEE</title></head>
+        <body>
+            <h1>Welcome to NORTHERN_AI</h1>
+            <form id="chat-form">
+                <input type="text" id="user-input" placeholder="Ask NORTHERN_AI something...">
+                <button type="submit">Send</button>
+            </form>
+            <div id="chat-history"></div>
+            
+            <script>
+                document.getElementById('chat-form').addEventListener('submit', async (e) => {
+                    e.preventDefault();
+                    const input = document.getElementById('user-input');
+                    const message = input.value;
+                    input.value = '';
+                    
+                    // Display user message
+                    const chatHistory = document.getElementById('chat-history');
+                    chatHistory.innerHTML += `<p><strong>You:</strong> ${message}</p>`;
+                    
+                    // Get AI response
+                    const response = await fetch('/api/chat', {
+                        method: 'POST',
+                        headers: {'Content-Type': 'application/json'},
+                        body: JSON.stringify({message})
+                    });
+                    
+                    const data = await response.json();
+                    chatHistory.innerHTML += `<p><strong>NORTHERN_AI:</strong> ${data.response}</p>`;
+                });
+            </script>
+        </body>
+    </html>
+    """
 
-if __name__ == "__main__":
-    demo.launch()
+if __name__ == '__main__':
+    # Use the PORT environment variable provided by most free hosting services
+    port = int(os.environ.get("PORT", 5000))
+    app.run(host='0.0.0.0', port=port)