Update app.py

app.py

@@ -84,7 +84,7 @@ custom_models = fetch_custom_models()
 all_models = ["huggingface", "groq", "mistral"] + custom_models
 
 # Determine the default model
-default_model = CUSTOM_LLM_DEFAULT_MODEL if CUSTOM_LLM_DEFAULT_MODEL in all_models else "huggingface"
+default_model = CUSTOM_LLM_DEFAULT_MODEL if CUSTOM_LLM_DEFAULT_MODEL in all_models else "groq"
 
 logger.info(f"Default model selected: {default_model}")
 
@@ -761,7 +761,7 @@ Remember to:
             response = groq_client.chat.completions.create(
                 messages=messages,
                 model="llama-3.1-70b-versatile",
-                max_tokens=500,
+                max_tokens=300,
                 temperature=temperature,
                 top_p=0.9,
                 presence_penalty=1.2,
@@ -773,7 +773,7 @@ Remember to:
             response = mistral_client.chat.complete(
                 model="open-mistral-nemo",
                 messages=messages,
-                max_tokens=500,
+                max_tokens=300,
                 temperature=temperature,
                 top_p=0.9,
                 stream=False
@@ -785,14 +785,14 @@ Remember to:
             custom_model = CustomModel(model)
             summary = custom_model.generate_response(
                 messages=messages,
-                max_tokens=500,
+                max_tokens=300,
                 temperature=temperature
             )
             
         else:  # huggingface
             response = client.chat_completion(
                 messages=messages,
-                max_tokens=500,
+                max_tokens=300,
                 temperature=temperature,
                 frequency_penalty=1.4,
                 top_p=0.9