Update test_phi3.py

test_phi3.py

@@ -1,65 +1,60 @@
-import streamlit as st
-from llama_cpp import Llama
-import os
-
-# ✅ Correct model path
-MODEL_PATH = "C:/models/Phi-3-mini-4k-instruct-q4.gguf"
-
-# ✅ Check if model exists
-if not os.path.exists(MODEL_PATH):
-    st.error(f"❌ Model not found: {MODEL_PATH}")
-    st.stop()
-
-# ✅ Load model ONCE in session
-if "model" not in st.session_state:
-    st.session_state["model"] = Llama(
-        model_path=MODEL_PATH,
-        n_ctx=4096,                 # Large context window
-        n_threads=os.cpu_count(),   # Use all CPU cores
-        n_batch=2048,               # Increase batch size for faster responses
-        use_mlock=True,             # Prevent memory swapping
-        n_kv_cache=128,             # Reduce KV Cache to save RAM
-        numa=False                  # Avoid NUMA issues
-    )
-
-st.title("🦙 Phi-3 Mini Chatbot")
-
-# ✅ User Input
-user_input = st.text_input("Ask me anything:", "")
-
-if st.button("Submit"):
-    if user_input:
-        st.write(f"**You:** {user_input}")
-
-        try:
-            # ✅ Generate response
-            response_data = st.session_state["model"].create_completion(
-                prompt=f"{user_input}\nAI:",
-                max_tokens=1024,  # Ensure full answers
-                temperature=0.7,
-                top_p=0.9,
-                stream=True
-            )
-
-            response_text = ""
-            response_container = st.empty()  # Placeholder for response
-
-            for chunk in response_data:
-                if "choices" in chunk and len(chunk["choices"]) > 0:
-                    choice = chunk["choices"][0]
-                    
-                    # ✅ Ensure valid text output
-                    if "text" in choice:
-                        response_text += choice["text"]
-                        response_container.markdown(f"**AI:** {response_text}")
-
-                    # ✅ Stop only when response is complete
-                    if choice.get("finish_reason") == "stop":
-                        break
-                else:
-                    st.warning("⚠️ Unexpected response format.")
-
-        except Exception as e:
-            st.error(f"🚨 Error generating response: {e}")
-
-
+import os
+import requests
+import streamlit as st
+from llama_cpp import Llama
+
+# ✅ Define model path inside Hugging Face Space
+MODEL_PATH = "./Phi-3-mini-4k-instruct-q4.gguf"
+MODEL_URL = "https://huggingface.co/YourModelRepo/Phi-3-mini-4k-instruct-q4.gguf"
+
+# ✅ Check if model exists, otherwise download
+if not os.path.exists(MODEL_PATH):
+    st.info("Downloading the model file. Please wait...")
+    with requests.get(MODEL_URL, stream=True) as response:
+        response.raise_for_status()
+        with open(MODEL_PATH, "wb") as f:
+            for chunk in response.iter_content(chunk_size=8192):
+                f.write(chunk)
+    st.success("Model downloaded successfully!")
+
+# ✅ Load model in session
+if "model" not in st.session_state:
+    st.session_state["model"] = Llama(model_path=MODEL_PATH, n_ctx=4096)
+
+# Streamlit UI setup
+st.set_page_config(page_title="Phi-3 Mini Chatbot", layout="centered")
+st.title("🤖 Phi-3 Mini Chatbot")
+st.markdown("Enter a message and get responses from Phi-3 Mini!")
+
+# Chat history
+if "messages" not in st.session_state:
+    st.session_state["messages"] = []
+
+# Display chat history
+for message in st.session_state["messages"]:
+    role, text = message
+    if role == "user":
+        st.chat_message("user").write(text)
+    else:
+        st.chat_message("assistant").write(text)
+
+# Input field for user message
+user_input = st.text_input("Your Message:", "", key="user_input")
+if st.button("Send") and user_input:
+    # Add user input to chat history
+    st.session_state["messages"].append(("user", user_input))
+    st.chat_message("user").write(user_input)
+    
+    # Generate response
+    response = st.session_state["model"].create_completion(
+        prompt=user_input, max_tokens=1024, temperature=0.7, top_p=0.9
+    )["choices"][0]["text"].strip()
+    
+    # Add model response to chat history
+    st.session_state["messages"].append(("assistant", response))
+    st.chat_message("assistant").write(response)
+
+# Run the app with: streamlit run test_phi3.py
+
+
+