Update app.py

app.py

@@ -117,7 +117,7 @@ def update_vectors(files, parser):
         label="Select documents to query"
     )
 
-def generate_chunked_response(prompt, model, max_tokens=1000, num_calls=3, temperature=0.2, should_stop=False):
+def generate_chunked_response(prompt, model, max_tokens=30000, num_calls=3, temperature=0.2, should_stop=False):
     print(f"Starting generate_chunked_response with {num_calls} calls")
     full_response = ""
     messages = [{"role": "user", "content": prompt}]
@@ -325,7 +325,8 @@ After writing the document, please provide a list of sources used in your respon
     payload = {
         "messages": inputs,
         "stream": True,
-        "temperature": temperature
+        "temperature": temperature,
+        "max_tokens": 32000
     }
 
     full_response = ""
@@ -376,7 +377,7 @@ After writing the document, please provide a list of sources used in your respon
         for i in range(num_calls):
             for message in client.chat_completion(
                 messages=[{"role": "user", "content": prompt}],
-                max_tokens=1000,
+                max_tokens=30000,
                 temperature=temperature,
                 stream=True,
             ):
@@ -437,7 +438,7 @@ Write a detailed and complete response that answers the following user question:
             logging.info(f"API call {i+1}/{num_calls}")
             for message in client.chat_completion(
                 messages=[{"role": "user", "content": prompt}],
-                max_tokens=1000,
+                max_tokens=30000,
                 temperature=temperature,
                 stream=True,
             ):