Update app.py

app.py

@@ -2,8 +2,9 @@ import os, time, json, re
 import gradio as gr
 from llama_cpp import Llama
 
+# ✅ Working public GGUF model
 MODEL_REPO = "QuantFactory/Phi-3.5-mini-instruct-GGUF"
-MODEL_FILE = "Phi-3.5-mini-instruct-Q4_K_M.gguf"
+MODEL_FILE = "Phi-3.5-mini-instruct.Q4_K_M.gguf"
 SAVE_PATH = "convos.jsonl"
 MAX_RECALL = 5  # how many past turns to recall
 
@@ -40,7 +41,6 @@ def load_memory(query):
         return []
     with open(SAVE_PATH, "r", encoding="utf-8") as f:
         lines = [json.loads(l) for l in f if l.strip()]
-    # naive keyword match
     query_words = set(re.findall(r"\w+", query.lower()))
     scored = []
     for rec in lines:
@@ -56,7 +56,6 @@ def format_messages(system, history, user_msg):
     if system.strip():
         msgs.append({"role": "system", "content": system})
 
-    # Inject recalled memory
     recalls = load_memory(user_msg)
     if recalls:
         mem_text = "\n".join(