|
from flask import Flask, request, jsonify, render_template |
|
from llama_cpp import Llama |
|
|
|
app = Flask(__name__) |
|
|
|
|
|
print("🔄 Loading model... (this may take a while)") |
|
llm = Llama.from_pretrained( |
|
repo_id="bartowski/google_gemma-3-1b-it-GGUF", |
|
filename="google_gemma-3-1b-it-IQ4_XS.gguf", |
|
) |
|
print("✅ Model loaded!") |
|
|
|
@app.route("/") |
|
def home(): |
|
print("📢 Serving index.html") |
|
return render_template("index.html") |
|
|
|
@app.route("/chat", methods=["POST"]) |
|
def chat(): |
|
user_input = request.json.get("message", "") |
|
print(f"💬 Received message: {user_input}") |
|
|
|
if not user_input: |
|
print("⚠️ Empty input received!") |
|
return jsonify({"error": "Empty input"}), 400 |
|
|
|
try: |
|
response = llm.create_chat_completion( |
|
messages=[{"role": "user", "content": user_input}] |
|
) |
|
|
|
print(f"🔍 Full response object: {response}") |
|
bot_reply = response["choices"][0]["message"]["content"] |
|
print(f"🤖 Bot response: {bot_reply}") |
|
|
|
return jsonify({"response": bot_reply}) |
|
|
|
except Exception as e: |
|
print(f"❌ Error generating response: {e}") |
|
return jsonify({"error": "Model failed to generate response"}), 500 |
|
|
|
if __name__ == "__main__": |
|
print("🚀 Starting Flask app on port 7860") |
|
app.run(host="0.0.0.0", port=7860, debug=True) |
|
|
|
|
|
|
