refactor: tokens

Author: wxg0103

apps/models_provider/impl/aliyun_bai_lian_model_provider/model/llm.py

@@ -20,5 +20,5 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
             model=model_name,
             openai_api_base=model_credential.get('api_base'),
             openai_api_key=model_credential.get('api_key'),
-            **optional_params
+            extra_body=optional_params
         )

apps/models_provider/impl/aws_bedrock_model_provider/model/llm.py

@@ -1,10 +1,12 @@
 import os
 import re
-from typing import Dict
+from typing import Dict, List
 
 from botocore.config import Config
 from langchain_community.chat_models import BedrockChat
+from langchain_core.messages import BaseMessage, get_buffer_string
 
+from common.config.tokenizer_manage_config import TokenizerManage
 from models_provider.base_model_provider import MaxKBBaseModel
 
 
@@ -72,6 +74,19 @@ def new_instance(cls, model_type: str, model_name: str, model_credential: Dict[s
             config=config
         )
 
+    def get_num_tokens_from_messages(self, messages: List[BaseMessage]) -> int:
+        try:
+            return super().get_num_tokens_from_messages(messages)
+        except Exception as e:
+            tokenizer = TokenizerManage.get_tokenizer()
+            return sum([len(tokenizer.encode(get_buffer_string([m]))) for m in messages])
+
+    def get_num_tokens(self, text: str) -> int:
+        try:
+            return super().get_num_tokens(text)
+        except Exception as e:
+            tokenizer = TokenizerManage.get_tokenizer()
+            return len(tokenizer.encode(text))
 
 def _update_aws_credentials(profile_name, access_key_id, secret_access_key):
     credentials_path = os.path.join(os.path.expanduser("~"), ".aws", "credentials")

apps/models_provider/impl/deepseek_model_provider/model/llm.py

@@ -26,6 +26,6 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
             model=model_name,
             openai_api_base='https://api.deepseek.com',
             openai_api_key=model_credential.get('api_key'),
-            **optional_params
+            extra_body=optional_params
         )
         return deepseek_chat_open_ai

apps/models_provider/impl/kimi_model_provider/model/llm.py

@@ -26,6 +26,6 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
             openai_api_base=model_credential['api_base'],
             openai_api_key=model_credential['api_key'],
             model_name=model_name,
-            **optional_params
+            extra_body=optional_params,
         )
         return kimi_chat_open_ai

apps/models_provider/impl/ollama_model_provider/credential/llm.py

@@ -25,7 +25,7 @@ class OllamaLLMModelParams(BaseForm):
                                     _step=0.01,
                                     precision=2)
 
-    max_tokens = forms.SliderField(
+    num_predict = forms.SliderField(
         TooltipLabel(_('Output the maximum Tokens'),
                      _('Specify the maximum number of tokens that the model can generate')),
         required=True, default_value=1024,

apps/models_provider/impl/openai_model_provider/model/llm.py

@@ -33,15 +33,15 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
         streaming = model_kwargs.get('streaming', True)
         if 'o1' in model_name:
             streaming = False
-        azure_chat_open_ai = OpenAIChatModel(
+        chat_open_ai = OpenAIChatModel(
             model=model_name,
             openai_api_base=model_credential.get('api_base'),
             openai_api_key=model_credential.get('api_key'),
-            **optional_params,
+            extra_body=optional_params,
             streaming=streaming,
             custom_get_token_ids=custom_get_token_ids
         )
-        return azure_chat_open_ai
+        return chat_open_ai
 
     def get_num_tokens_from_messages(self, messages: List[BaseMessage]) -> int:
         try:

apps/models_provider/impl/siliconCloud_model_provider/model/llm.py

@@ -34,5 +34,5 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
             model=model_name,
             openai_api_base=model_credential.get('api_base'),
             openai_api_key=model_credential.get('api_key'),
-            **optional_params
+            extra_body=optional_params
         )

apps/models_provider/impl/vllm_model_provider/model/llm.py

@@ -31,7 +31,7 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
             model=model_name,
             openai_api_base=model_credential.get('api_base'),
             openai_api_key=model_credential.get('api_key'),
-            **optional_params,
+            extra_body=optional_params,
             streaming=True,
             stream_usage=True,
         )

apps/models_provider/impl/volcanic_engine_model_provider/model/llm.py

@@ -17,5 +17,5 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
             model=model_name,
             openai_api_base=model_credential.get('api_base'),
             openai_api_key=model_credential.get('api_key'),
-            **optional_params
+            extra_body=optional_params
         )

apps/models_provider/impl/xinference_model_provider/model/llm.py

@@ -34,7 +34,7 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
             model=model_name,
             openai_api_base=base_url,
             openai_api_key=model_credential.get('api_key'),
-            **optional_params
+            extra_body=optional_params
         )
 
     def get_num_tokens_from_messages(self, messages: List[BaseMessage]) -> int: