fix: compatibility issues with the tongyi model. (#1310)

api/core/model_providers/models/llm/tongyi_model.py

@@ -18,7 +18,6 @@ class TongyiModel(BaseLLM):
 
     def _init_client(self) -> Any:
         provider_model_kwargs = self._to_model_kwargs_input(self.model_rules, self.model_kwargs)
-        del provider_model_kwargs['max_tokens']
         return EnhanceTongyi(
             model_name=self.name,
             max_retries=1,
@@ -58,7 +57,6 @@ class TongyiModel(BaseLLM):
 
     def _set_model_kwargs(self, model_kwargs: ModelKwargs):
         provider_model_kwargs = self._to_model_kwargs_input(self.model_rules, model_kwargs)
-        del provider_model_kwargs['max_tokens']
         for k, v in provider_model_kwargs.items():
             if hasattr(self.client, k):
                 setattr(self.client, k, v)

api/core/model_providers/providers/tongyi_provider.py

@@ -24,12 +24,12 @@ class TongyiProvider(BaseModelProvider):
         if model_type == ModelType.TEXT_GENERATION:
             return [
                 {
-                    'id': 'qwen-v1',
-                    'name': 'qwen-v1',
+                    'id': 'qwen-turbo',
+                    'name': 'qwen-turbo',
                 },
                 {
-                    'id': 'qwen-plus-v1',
-                    'name': 'qwen-plus-v1',
+                    'id': 'qwen-plus',
+                    'name': 'qwen-plus',
                 }
             ]
         else:
@@ -58,16 +58,16 @@ class TongyiProvider(BaseModelProvider):
         :return:
         """
         model_max_tokens = {
-            'qwen-v1': 1500,
-            'qwen-plus-v1': 6500
+            'qwen-turbo': 6000,
+            'qwen-plus': 6000
         }
 
         return ModelKwargsRules(
-            temperature=KwargRule[float](enabled=False),
-            top_p=KwargRule[float](min=0, max=1, default=0.8, precision=2),
+            temperature=KwargRule[float](min=0.01, max=1, default=1, precision=2),
+            top_p=KwargRule[float](min=0.01, max=0.99, default=0.5, precision=2),
             presence_penalty=KwargRule[float](enabled=False),
             frequency_penalty=KwargRule[float](enabled=False),
-            max_tokens=KwargRule[int](min=10, max=model_max_tokens.get(model_name), default=1024, precision=0),
+            max_tokens=KwargRule[int](enabled=False, max=model_max_tokens.get(model_name)),
         )
 
     @classmethod
@@ -84,7 +84,7 @@ class TongyiProvider(BaseModelProvider):
             }
 
             llm = EnhanceTongyi(
-                model_name='qwen-v1',
+                model_name='qwen-turbo',
                 max_retries=1,
                 **credential_kwargs
             )

api/core/model_providers/rules/tongyi.json

@@ -3,5 +3,19 @@
         "custom"
     ],
     "system_config": null,
-    "model_flexibility": "fixed"
+    "model_flexibility": "fixed",
+    "price_config": {
+        "qwen-turbo": {
+            "prompt": "0.012",
+            "completion": "0.012",
+            "unit": "0.001",
+            "currency": "RMB"
+        },
+        "qwen-plus": {
+            "prompt": "0.14",
+            "completion": "0.14",
+            "unit": "0.001",
+            "currency": "RMB"
+        }
+    }
 }

api/requirements.txt

@@ -44,7 +44,7 @@ readabilipy==0.2.0
 google-search-results==2.4.2
 replicate~=0.9.0
 websocket-client~=1.6.1
-dashscope~=1.5.0
+dashscope~=1.11.0
 huggingface_hub~=0.16.4
 transformers~=4.31.0
 stripe~=5.5.0

api/tests/integration_tests/models/llm/test_tongyi_model.py

@@ -44,7 +44,7 @@ def decrypt_side_effect(tenant_id, encrypted_api_key):
 
 @patch('core.helper.encrypter.decrypt_token', side_effect=decrypt_side_effect)
 def test_get_num_tokens(mock_decrypt):
-    model = get_mock_model('qwen-v1')
+    model = get_mock_model('qwen-turbo')
     rst = model.get_num_tokens([
         PromptMessage(type=MessageType.HUMAN, content='Who is your manufacturer?')
     ])
@@ -55,7 +55,7 @@ def test_get_num_tokens(mock_decrypt):
 def test_run(mock_decrypt, mocker):
     mocker.patch('core.model_providers.providers.base.BaseModelProvider.update_last_used', return_value=None)
 
-    model = get_mock_model('qwen-v1')
+    model = get_mock_model('qwen-turbo')
     rst = model.run(
         [PromptMessage(content='Human: Are you Human? you MUST only answer `y` or `n`? \nAssistant: ')],
         stop=['\nHuman:'],