Update app.py

app.py

@@ -1,49 +1,30 @@
-import os
-import streamlit as st
-from llama_cpp import Llama
 from huggingface_hub import hf_hub_download
+from llama_cpp import Llama
+import streamlit as st
 
-# Hugging Face repo and model info
+# Model info
 REPO_ID = "MaziyarPanahi/BioMistral-7B-GGUF"
-MODEL_FILENAME = "BioMistral-7B.Q4_K_M.gguf"
+FILENAME = "BioMistral-7B.Q4_K_M.gguf"
 
-# Streamlit UI
 st.set_page_config(page_title="Medical Chatbot")
 st.title("🩺 Medical Chatbot using BioMistral-7B")
 
-# Download model via huggingface_hub
-with st.spinner("🔄 Checking/downloading model (this may take a few minutes)..."):
-    try:
-        model_path = hf_hub_download(
-            repo_id=REPO_ID,
-            filename=MODEL_FILENAME,
-            cache_dir="models"  # Optional: use your preferred directory
-        )
-    except Exception as e:
-        st.error(f"❌ Failed to download model: {e}")
-        st.stop()
+with st.spinner("📦 Downloading model (first-time only)..."):
+    model_path = hf_hub_download(repo_id=REPO_ID, filename=FILENAME)
 
 # Load model
 with st.spinner("⚙️ Loading BioMistral model..."):
-    try:
-        llm = Llama(
-            model_path=model_path,
-            n_ctx=4096,
-            n_threads=8,
-            n_gpu_layers=35  # Set to 0 for CPU
-        )
-    except Exception as e:
-        st.error(f"❌ Failed to load model: {e}")
-        st.stop()
+    llm = Llama(
+        model_path=model_path,
+        n_ctx=4096,
+        n_threads=8,
+        n_gpu_layers=35  # Adjust to 0 if CPU-only
+    )
 
-# Chat UI
-query = st.text_input("💬 Ask a medical question:")
+# UI
+query = st.text_input("💬 Enter your medical question:")
 if query:
-    with st.spinner("🧠 Generating answer..."):
-        try:
-            response = llm(query, max_tokens=512, stop=["</s>"])
-            answer = response["choices"][0]["text"].strip()
-            st.markdown("**🩺 Answer:**")
-            st.write(answer)
-        except Exception as e:
-            st.error(f"⚠️ Error generating response: {e}")
+    with st.spinner("🧠 Thinking..."):
+        response = llm(query, max_tokens=512, stop=["</s>"])
+        st.markdown("**Answer:**")
+        st.write(response["choices"][0]["text"].strip())