Update app.py

app.py

@@ -1,19 +1,22 @@
 import os
 import torch
 import datetime
-from flask import Flask, request, Response
+from flask import Flask, request, jsonify, Response
 from flask_cors import CORS
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 from threading import Thread
 
 app = Flask(__name__)
-# 🔱 ඉතා වැදගත්: පිටතින් එන HTML එකට සම්බන්ධ වීමට CORS අනිවාර්යයි
-CORS(app)
+# 🔱 වැදගත්: ඕනෑම තැනක සිට සම්බන්ධ වීමට CORS ඉඩ ලබා දීම
+CORS(app, resources={r"/api/*": {"origins": "*"}})
 
+# ============================================
+# 🔱 TRANSFORMERS ENGINE CONFIG
+# ============================================
 MODEL_ID = "Qwen/Qwen2.5-1.5B-Instruct"
 HF_TOKEN = os.getenv("HF_TOKEN")
 
-print("🐘 Elephant Core Backend is booting up...")
+print("🐘 Helpful Elephant is loading via Transformers...")
 
 tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, token=HF_TOKEN)
 model = AutoModelForCausalLM.from_pretrained(
@@ -23,30 +26,54 @@ model = AutoModelForCausalLM.from_pretrained(
     token=HF_TOKEN
 )
 
-SYSTEM_PROMPT = "🐘 HELPFUL ELEPHANT AI. Authorized by MINZO-PRIME. Provide technical precision."
+SYSTEM_PROMPT = "🐘 HELPFUL ELEPHANT AI v1.0. High-speed Research Mode. Commander: MINZO-PRIME."
 
+# ============================================
+# 🔱 API ENDPOINT
+# ============================================
 @app.route('/api/chat', methods=['POST'])
 def chat():
-    data = request.json
-    user_msg = data.get('message', '').strip()
-    
-    messages = [
-        {"role": "system", "content": SYSTEM_PROMPT},
-        {"role": "user", "content": user_msg}
-    ]
-    
-    inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
-    streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
-    
-    gen_kwargs = dict(input_ids=inputs, streamer=streamer, max_new_tokens=1024, temperature=0.7, do_sample=True)
-    thread = Thread(target=model.generate, kwargs=gen_kwargs)
-    thread.start()
+    try:
+        data = request.json
+        user_msg = data.get('message', '').strip()
+        
+        messages = [
+            {"role": "system", "content": SYSTEM_PROMPT},
+            {"role": "user", "content": user_msg}
+        ]
+        
+        inputs = tokenizer.apply_chat_template(
+            messages,
+            add_generation_prompt=True,
+            return_tensors="pt"
+        ).to(model.device)
+
+        streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
+        
+        generation_kwargs = dict(
+            input_ids=inputs,
+            streamer=streamer,
+            max_new_tokens=1024,
+            temperature=0.7,
+            do_sample=True
+        )
 
-    def generate():
-        for new_text in streamer:
-            yield new_text
+        thread = Thread(target=model.generate, kwargs=generation_kwargs)
+        thread.start()
+
+        def generate():
+            for new_text in streamer:
+                yield new_text
+
+        return Response(generate(), mimetype='text/plain')
+    
+    except Exception as e:
+        return jsonify({"error": str(e)}), 500
 
-    return Response(generate(), mimetype='text/plain')
+@app.route('/')
+def health():
+    return "🐘 Elephant AI Core is Online. Port: 7860"
 
 if __name__ == '__main__':
+    # 🔱 Hugging Face සඳහා අනිවාර්යයෙන්ම port 7860 විය යුතුයි
     app.run(host='0.0.0.0', port=7860)