added default llm 4o mini and optional limit if pass from payload in …

…db action (#1373) * 1. added gpt-4o-mini as default 2. optional limit parameter in qdrant if provided by pauload * 1. added gpt-4o-mini as default 2. optional limit parameter in qdrant if provided by pauload * 1. added gpt-4o-mini as default 2. optional limit parameter in qdrant if provided by pauload
digiteinfotech · Jul 25, 2024 · 2006317 · 2006317
1 parent 7659149
commit 2006317
Show file tree

Hide file tree

Showing 9 changed files with 65 additions and 63 deletions.
diff --git a/kairon/shared/vector_embeddings/db/qdrant.py b/kairon/shared/vector_embeddings/db/qdrant.py
@@ -45,7 +45,9 @@ async def perform_operation(self, data: Dict, user: str, **kwargs):
                 request.update(**payload)
 
         if request:
-            request.update(**{'with_payload': True, 'limit': 10})
+            request.update(**{'with_payload': True})
+            if 'limit' not in request:
+                request['limit'] = 10
             result = ActionUtility.execute_http_request(http_url=url,
                                                         request_method='POST',
                                                         request_body=request)

diff --git a/metadata/integrations.yml b/metadata/integrations.yml
@@ -114,7 +114,7 @@ llm:
         description: "The max_tokens hyperparameter limits the length of generated responses in chat completion using ChatGPT."
       model:
         type: string
-        default: "gpt-3.5-turbo"
+        default: "gpt-4o-mini"
         enum: ["gpt-3.5-turbo", "gpt-3.5-turbo-instruct", "gpt-4o-mini"]
         description: "The model hyperparameter is the ID of the model to use such as gpt-2, gpt-3, or a custom model that you have trained or fine-tuned."
       top_p:

diff --git a/template/use-cases/Hi-Hello-GPT/actions.yml b/template/use-cases/Hi-Hello-GPT/actions.yml
@@ -18,7 +18,7 @@ prompt_action:
     frequency_penalty: 0
     logit_bias: {}
     max_tokens: 300
-    model: gpt-3.5-turbo
+    model: gpt-4o-mini
     n: 1
     presence_penalty: 0
     stop: null

diff --git a/tests/integration_test/action_service_test.py b/tests/integration_test/action_service_test.py
diff --git a/tests/integration_test/services_test.py b/tests/integration_test/services_test.py
@@ -4840,7 +4840,7 @@ def test_get_prompt_action():
         {'name': 'test_update_prompt_action', 'num_bot_responses': 5, 'failure_message': 'updated_failure_message',
          'user_question': {'type': 'from_slot', 'value': 'prompt_question'},
          'llm_type': 'openai',
-         'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-3.5-turbo', 'top_p': 0.0, 'n': 1,
+         'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-4o-mini', 'top_p': 0.0, 'n': 1,
                              'stop': None, 'presence_penalty': 0.0, 'frequency_penalty': 0.0,
                              'logit_bias': {}}, 'llm_prompts': [
             {'name': 'System Prompt', 'data': 'You are a personal assistant.', 'type': 'system', 'source': 'static',
@@ -4935,7 +4935,7 @@ def _mock_get_bot_settings(*args, **kwargs):
         'failure_message': "I'm sorry, I didn't quite understand that. Could you rephrase?",
         'user_question': {'type': 'from_user_message'},
         'llm_type': 'openai',
-        'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-3.5-turbo', 'top_p': 0.0, 'n': 1,
+        'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-4o-mini', 'top_p': 0.0, 'n': 1,
                             'stop': None, 'presence_penalty': 0.0, 'frequency_penalty': 0.0,
                             'logit_bias': {}},
         'llm_prompts': [{'name': 'System Prompt', 'data': 'You are a personal assistant.', 'type': 'system',
@@ -5027,7 +5027,7 @@ def _mock_get_bot_settings(*args, **kwargs):
         'failure_message': "I'm sorry, I didn't quite understand that. Could you rephrase?",
         'user_question': {'type': 'from_user_message'},
         'llm_type': 'openai',
-        'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-3.5-turbo', 'top_p': 0.0, 'n': 1,
+        'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-4o-mini', 'top_p': 0.0, 'n': 1,
                             'stop': None, 'presence_penalty': 0.0,
                             'frequency_penalty': 0.0, 'logit_bias': {}},
         'llm_prompts': [{'name': 'System Prompt', 'data': 'You are a personal assistant.', 'type': 'system',
@@ -5122,7 +5122,7 @@ def _mock_get_bot_settings(*args, **kwargs):
         'failure_message': "I'm sorry, I didn't quite understand that. Could you rephrase?",
         'user_question': {'type': 'from_user_message'},
         'llm_type': 'openai',
-        'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-3.5-turbo', 'top_p': 0.0, 'n': 1,
+        'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-4o-mini', 'top_p': 0.0, 'n': 1,
                             'stop': None, 'presence_penalty': 0.0, 'frequency_penalty': 0.0,
                             'logit_bias': {}},
         'llm_prompts': [{'name': 'System Prompt', 'data': 'You are a personal assistant.', 'type': 'system',
@@ -24975,7 +24975,7 @@ def test_get_llm_logs():
     expected = "Hi, How may i help you?"
 
     result = loop.run_until_complete(litellm.acompletion(messages=messages,
-                                                         model="gpt-3.5-turbo",
+                                                         model="gpt-4o-mini",
                                                          mock_response=expected,
                                                          metadata={'user': user, 'bot': pytest.bot}))
     assert result['choices'][0]['message']['content'] == expected

diff --git a/tests/unit_test/action/action_test.py b/tests/unit_test/action/action_test.py
@@ -2662,7 +2662,7 @@ def test_get_prompt_action_config(self):
         assert actual_config == {'name': 'kairon_faq_action', 'num_bot_responses': 5,
                                  'failure_message': "I'm sorry, I didn't quite understand that. Could you rephrase?",
                                  'user_question': {'type': 'from_user_message'}, 'bot': 'test_action_server',
-                                 'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-3.5-turbo',
+                                 'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-4o-mini',
                                                      'top_p': 0.0, 'n': 1, 'stop': None,
                                                      'presence_penalty': 0.0, 'frequency_penalty': 0.0,
                                                      'logit_bias': {}},
@@ -3956,7 +3956,7 @@ def test_get_prompt_action_config_2(self):
                                        'user_question': {'type': 'from_user_message'},
                                        'failure_message': "I'm sorry, I didn't quite understand that. Could you rephrase?",
                                        'bot': 'test_bot_action_test', 'user': 'test_user_action_test',
-                                       'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-3.5-turbo',
+                                       'hyperparameters': {'temperature': 0.0, 'max_tokens': 300, 'model': 'gpt-4o-mini',
                                                            'top_p': 0.0, 'n': 1, 'stop': None,
                                                            'presence_penalty': 0.0, 'frequency_penalty': 0.0,
                                                            'logit_bias': {}}, 'dispatch_response': True, 'set_slots': [],