Update app.py

app.py

@@ -483,30 +483,6 @@ You are an expert AI assistant. Write a detailed summary of the information prov
 Base your summary strictly on the information from this source. Only include information that is directly supported by the given content.
 If any part of the information cannot be verified from this source, clearly state that it could not be confirmed."""
 
-        if model == "@cf/meta/llama-3.1-8b-instruct":
-            # Use Cloudflare API
-            source_response = ""
-            for response in get_response_from_cloudflare(prompt="", context=context, query=query, num_calls=1, temperature=temperature, search_type="web"):
-                source_response += response
-            accumulated_response += f"Source {i} ({source}):\n\n{source_response}\n\n"
-            yield accumulated_response, ""
-        else:
-            # Use Hugging Face API
-            client = InferenceClient(model, token=huggingface_token)
-            
-            source_response = ""
-            for message in client.chat_completion(
-                messages=[{"role": "user", "content": prompt}],
-                max_tokens=2000,
-                temperature=temperature,
-                stream=True,
-            ):
-                if message.choices and message.choices[0].delta and message.choices[0].delta.content:
-                    chunk = message.choices[0].delta.content
-                    source_response += chunk
-            accumulated_response += f"Source {i} ({source}):\n\n{source_response}\n\n"
-            yield accumulated_response, ""
-    
     # Generate an overall summary after processing all sources
     overall_prompt = f"""Based on the summaries you've generated for each source: '{accumulated_response}', provide a concise overall summary that addresses the user's query: '{query}'
 Highlight any conflicting information or gaps in the available data."""
@@ -523,7 +499,7 @@ Highlight any conflicting information or gaps in the available data."""
         overall_summary = ""
         for message in client.chat_completion(
             messages=[{"role": "user", "content": overall_prompt}],
-            max_tokens=2000,
+            max_tokens=10000,
             temperature=temperature,
             stream=True,
         ):
@@ -533,7 +509,6 @@ Highlight any conflicting information or gaps in the available data."""
         accumulated_response += f"Overall Summary:\n\n{overall_summary}\n\n"
         yield accumulated_response, ""
 
-
 def get_response_from_pdf(query, model, selected_docs, num_calls=3, temperature=0.2):
     logging.info(f"Entering get_response_from_pdf with query: {query}, model: {model}, selected_docs: {selected_docs}")