Update app.py

app.py

@@ -68,6 +68,7 @@ if "chunks_store" not in st.session_state:
 index = st.session_state.faiss_index
 chunks_store = st.session_state.chunks_store
 
+
 # ------------------------- Helpers -------------------------
 def chunk_text(text: str, max_length: int = 500):
     words, chunks, cur = text.split(), [], []
@@ -91,24 +92,20 @@ def embed_and_store(chunks):
 
 
 def query_llm(prompt: str) -> str:
-    """Stream a concise response from Groq and return full text."""
+    """Non-streaming version for stability on Hugging Face."""
     try:
-        stream = client.chat.completions.create(
+        completion = client.chat.completions.create(
             model="deepseek-r1-distill-llama-70b",
             messages=[
                 {"role": "system", "content": SYSTEM_MSG},
                 {"role": "user", "content": prompt},
             ],
-            temperature=0.4,
-            max_completion_tokens=220,
+            temperature=0.5,
+            max_completion_tokens=300,
             top_p=0.9,
-            stream=True,
-            reasoning_format="hidden",  # valid values: 'hidden', 'raw', or 'parsed'
         )
-        out = []
-        for chunk in stream:
-            out.append(chunk.choices[0].delta.content or "")
-        return "".join(out)
+        # Return final text directly
+        return completion.choices[0].message.content.strip()
     except Exception as e:
         return f"⚠️ **Error:** {str(e)}"
 
@@ -140,7 +137,7 @@ if uploaded_file:
             context = " ".join(relevant)
             final_prompt = f"Context:\n{context}\n\nQuestion:\n{user_query}"
 
-            with st.spinner("Analyzing…"):
+            with st.spinner("Analyzing… please wait."):
                 answer = query_llm(final_prompt)
 
             st.markdown("### AI Analysis")