swarmauri 0.3.0.dev37__tar.gz → 0.3.2.dev1__tar.gz

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: swarmauri
-Version: 0.3.0.dev37
+Version: 0.3.2.dev1
 Summary: This repository includes core interfaces, standard ABCs and concrete references, third party plugins, and experimental modules for the swarmaURI framework.
 Home-page: http://github.com/swarmauri/swarmauri-sdk
 Author: Jacob Stewart

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.3.0.dev37"
+__version__ = "0.3.2.dev1"
 __long_desc__ = """
 # swarmaURI sdk
 

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/core/models/IPredict.py

@@ -6,7 +6,7 @@ class IPredict(ABC):
     """
 
     @abstractmethod
-    def predict(self, input_data) -> any:
+    def predict(self, *args, **kwargs) -> any:
         """
         Generate predictions based on the input data provided to the model.
         """

swarmauri-0.3.2.dev1/swarmauri/standard/models/base/ModelBase.py

@@ -0,0 +1,29 @@
+from abc import ABC, abstractmethod
+from typing import Any, Union, Optional
+from dataclasses import dataclass, field
+from swarmauri.core.models.IModel import IModel
+from swarmauri.core.models.IPredict import IPredict
+from swarmauri.core.BaseComponent import BaseComponent, ResourceTypes
+
+@dataclass
+class ModelBase(IPredict, IModel, BaseComponent):
+    model_name: Optional[str] = None
+    resource: Optional[str] =  field(default=ResourceTypes.MODEL.value)
+
+    def __post_init__(self):
+        if not self.model_name:
+            raise ValueError('Must define model_name.')
+            
+    @property
+    def model_name(self):
+        return self._model_name
+    
+    @model_name.setter
+    def model_name(self, value: Union[str, None]) -> None:
+        """
+        Property setter that sets the name of the model.
+
+        Parameters:
+        - value (str): The new name of the model.
+        """
+        self._model_name = value

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/models/concrete/AnthropicModel.py

@@ -1,25 +1,28 @@
 import json
 from typing import List
+from dataclasses import dataclass
 import anthropic
-import logging
-from swarmauri.core.models.IPredict import IPredict
 from swarmauri.standard.models.base.ModelBase import ModelBase
 
-class AnthropicModel(ModelBase, IPredict):
+@dataclass
+class AnthropicModel(ModelBase):
     allowed_models = ['claude-3-opus-20240229', 
     'claude-3-sonnet-20240229', 
     'claude-3-haiku-20240307',
     'claude-2.1',
     'claude-2.0',
     'claude-instant-1.2']
+    api_key: str = ""
+    model_name: str = "claude-3-haiku-20240307"
+
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client = anthropic.Anthropic(api_key=self.api_key)
+
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
 
-    def __init__(self, api_key: str, model_name: str = 'claude-3-haiku-20240307'):
-        if model_name not in self.allowed_models:
-            raise ValueError(f"Model name '{model_name}' is not supported. Choose from {self.allowed_models}")
-        
-        self.client = anthropic.Anthropic(api_key=api_key)
-        super().__init__(model_name)
-        
     
     def predict(self, messages, temperature=0.7, max_tokens=256):
 
@@ -28,7 +31,6 @@ class AnthropicModel(ModelBase, IPredict):
         for message in messages:
             if message['role'] == 'system':
                 system_context = message['content']
-                logging.info(f'Setting system to {system_context}.')
 
         
         # Remove system instruction from messages
@@ -51,4 +53,4 @@ class AnthropicModel(ModelBase, IPredict):
             )
         
         message_content = response.content[0].text
-        return message_content
+        return message_content

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/models/concrete/CohereModel.py

@@ -1,23 +1,26 @@
 import json
 from typing import List
+from dataclasses import dataclass
 import cohere
-from swarmauri.core.models.IPredict import IPredict
 from swarmauri.standard.models.base.ModelBase import ModelBase
 
-
-class CohereModel(ModelBase, IPredict):
+@dataclass
+class CohereModel(ModelBase):
     allowed_models = ['command-light',
     'command', 
     'command-r',
     'command-r-plus']
+    api_key: str = ""
+    model_name: str = "command-light"
+
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client = cohere.Client(api_key=self.api_key)
+
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
 
-    def __init__(self, api_key: str, model_name: str = 'command-light'):
-        if model_name not in self.allowed_models:
-            raise ValueError(f"Model name '{model_name}' is not supported. Choose from {self.allowed_models}")
-        
-        self.client = cohere.Client(api_key=api_key)
-        super().__init__(model_name)
-        
     
     def predict(self, messages, temperature=0.7, max_tokens=256):
         response = self.client.chat(

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/models/concrete/GeminiProModel.py

@@ -1,18 +1,19 @@
 import json
 from typing import List
+from dataclasses import dataclass
 import google.generativeai as genai
-from swarmauri.core.models.IPredict import IPredict
 from swarmauri.standard.models.base.ModelBase import ModelBase
 
-
-class GeminiProModel(ModelBase, IPredict):
+@dataclass
+class GeminiProModel(ModelBase):
     allowed_models = ['gemini-1.5-pro-latest']
+    api_key: str = ""
+    model_name: str = "gemini-1.5-pro-latest"
 
-    def __init__(self, api_key: str, model_name: str = 'gemini-1.5-pro-latest'):
-        if model_name not in self.allowed_models:
-            raise ValueError(f"Model name '{model_name}' is not supported. Choose from {self.allowed_models}")
-        
-        genai.configure(api_key=api_key)
+
+    def __post_init__(self):
+        self._validate_model_name()
+        genai.configure(api_key=self.api_key)
         self.safety_settings = [
           {
             "category": "HARM_CATEGORY_HARASSMENT",
@@ -50,8 +51,11 @@ class GeminiProModel(ModelBase, IPredict):
           },
         ]
         self.client = None
-        super().__init__(model_name)
-        
+
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
+
     
     def predict(self, messages, temperature=0.7, max_tokens=256):
         generation_config = {

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/models/concrete/GroqModel.py

@@ -1,20 +1,25 @@
 import json
 from typing import List
+from dataclasses import dataclass
 from groq import Groq
-from swarmauri.core.models.IPredict import IPredict
 from swarmauri.standard.models.base.ModelBase import ModelBase
 
+@dataclass
+class GroqModel(ModelBase):
+    allowed_models = ['llama3-8b-8192', 
+    'llama3-70b-8192', 
+    'mixtral-8x7b-32768', 
+    'gemma-7b-it']
+    api_key: str = ""
+    model_name: str = "mixtral-8x7b-32768"
 
-class GroqModel(ModelBase, IPredict):
-    allowed_models = ['llama3-8b-8192', 'llama3-70b-8192', 'mixtral-8x7b-32768', 'gemma-7b-it']
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client = Groq(api_key=self.api_key)
 
-    def __init__(self, api_key: str, model_name: str = 'mixtral-8x7b-32768'):
-        if model_name not in self.allowed_models:
-            raise ValueError(f"Model name '{model_name}' is not supported. Choose from {self.allowed_models}")
-        
-        self.client = Groq(api_key=api_key)
-        super().__init__(model_name)
-        
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
     
     def predict(self, messages, 
         temperature=0.7, 

swarmauri-0.3.2.dev1/swarmauri/standard/models/concrete/GroqToolModel.py

@@ -0,0 +1,35 @@
+from dataclasses import dataclass
+from groq import Groq
+from swarmauri.standard.models.base.ModelBase import ModelBase
+
+@dataclass
+class GroqToolModel(ModelBase):
+    allowed_models = ['llama3-8b-8192', 
+    'llama3-70b-8192', 
+    'mixtral-8x7b-32768', 
+    'gemma-7b-it']
+    api_key: str = ""
+    model_name: str = "mixtral-8x7b-32768"
+
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client = Groq(api_key=self.api_key)
+
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
+    
+        
+
+    def predict(self, messages, tools=None, tool_choice=None, temperature=0.7, max_tokens=1024):
+        if tools and not tool_choice:
+            tool_choice = "auto"
+        response = self.client.chat.completions.create(
+            model=self.model_name,
+            messages=messages,
+            temperature=temperature,
+            max_tokens=max_tokens,
+            tools=tools,
+            tool_choice=tool_choice,
+        )
+        return response

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/models/concrete/MistralModel.py

@@ -1,27 +1,30 @@
 import json
 from typing import List
+from dataclasses import dataclass
 from mistralai.client import MistralClient
-from swarmauri.core.models.IPredict import IPredict
 from swarmauri.standard.models.base.ModelBase import ModelBase
 
-
-class MistralModel(ModelBase, IPredict):
+class MistralModel(ModelBase):
     allowed_models = ['open-mistral-7b', 
     'open-mixtral-8x7b', 
     'open-mixtral-8x22b', 
     'mistral-small-latest',
     'mistral-medium-latest',
     'mistral-large-latest',
+    'codestral'
     ]
+    api_key: str = ""
+    model_name: str = "open-mixtral-8x7b"
+
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client =  MistralClient(api_key=self.api_key)
+
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
+
 
-    def __init__(self, api_key: str, model_name: str = 'open-mixtral-8x7b'):
-        if model_name not in self.allowed_models:
-            raise ValueError(f"Model name '{model_name}' is not supported. Choose from {self.allowed_models}")
-        
-        self.client = MistralClient(api_key=api_key)
-        super().__init__(model_name)
-        
-    
     def predict(self, messages, 
         temperature: int = 0.7, 
         max_tokens: int = 256, 

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/models/concrete/MistralToolModel.py

@@ -1,19 +1,24 @@
+import json
+from typing import List
+from dataclasses import dataclass
 from mistralai.client import MistralClient
 from swarmauri.standard.models.base.ModelBase import ModelBase
-from swarmauri.core.models.IPredict import IPredict
 
-class MistralToolModel(ModelBase, IPredict):
+class MistralModel(ModelBase):
     allowed_models = ['open-mixtral-8x22b', 
     'mistral-small-latest',
     'mistral-large-latest',
     ]
+    api_key: str = ""
+    model_name: str = "open-mixtral-8x22b"
 
-    def __init__(self, api_key: str, model_name: str = 'open-mixtral-8x22b'):
-        if model_name not in self.allowed_models:
-            raise ValueError(f"Model name '{model_name}' is not supported. Choose from {self.allowed_models}")
-        
-        self.client = MistralClient(api_key=api_key)
-        super().__init__(model_name)
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client =  MistralClient(api_key=self.api_key)
+
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
         
 
     def predict(self, messages, tools=None, tool_choice=None, temperature=0.7, 

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/models/concrete/OpenAIImageGenerator.py

@@ -1,20 +1,21 @@
 import json
+from dataclasses import dataclass
 from openai import OpenAI
-from ..base.ModelBase import ModelBase
-from ....core.models.IPredict import IPredict
+from swarmauri.core.models.base.ModelBase import ModelBase
 
-class OpenAIImageGenerator(ModelBase, IPredict):
-    def __init__(self, api_key: str, model_name: str = "dall-e"):
-        """
-        Initializes the OpenAI image generator model.
+@dataclass
+class OpenAIImageGenerator(ModelBase):
+    allowed_models = ['dall-e']
+    api_key: str = ""
+    model_name: str = "dall-e"
 
-        Parameters:
-        - api_key (str): The API key provided by OpenAI for access to their services.
-        - model_name (str): Name of the image generation model provided by OpenAI.
-                            Defaults to "dall-e" for DALL·E, their image generation model.
-        """
-        self.client = OpenAI(api_key=api_key)
-        super().__init__(model_name)
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client =  OpenAI(api_key=self.api_key)
+
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
 
     def predict(self, prompt: str, size: str = "1024x1024", 
                 quality: str = "standard", n: int = 1) -> str:

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/models/concrete/OpenAIModel.py

@@ -1,36 +1,38 @@
 import json
 from typing import List
+from dataclasses import dataclass
 from openai import OpenAI
-from swarmauri.core.models.IPredict import IPredict
 from swarmauri.standard.models.base.ModelBase import ModelBase
 
-
-class OpenAIModel(ModelBase, IPredict):
+@dataclass
+class OpenAIModel(ModelBase):
     allowed_models = ['gpt-4o', 
+    'gpt-4o-2024-05-13',
     'gpt-4-turbo', 
+    'gpt-4-turbo-2024-04-09',
+    'gpt-4-turbo-preview',
     'gpt-4-0125-preview',
+    'gpt-4-1106-preview',
     'gpt-4',
     'gpt-4-0613',
     'gpt-4-32k',
     'gpt-4-32k-0613',
     'gpt-3.5-turbo-0125',
+    'gpt-3.5-turbo-1106',
+    'gpt-3.5-turbo-0613',
     'gpt-3.5-turbo-16k-0613',
     'gpt-3.5-turbo-16k',
     'gpt-3.5-turbo']
+    api_key: str = ""
+    model_name: str = "gpt-3.5-turbo-16k"
 
-    def __init__(self, api_key: str, model_name: str):
-        """
-        Initialize the OpenAI model with an API key.
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client = OpenAI(api_key=self.api_key)
 
-        Parameters:
-        - api_key (str): Your OpenAI API key.
-        """
-        if model_name not in self.allowed_models:
-            raise ValueError(f"Model name '{model_name}' is not supported. Choose from {self.allowed_models}")
-        
-        self.client = OpenAI(api_key=api_key)
-        super().__init__(model_name)
-        
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
     
     def predict(self, messages, temperature=0.7, max_tokens=256, enable_json=False, stop: List[str] = None):
         """

swarmauri-0.3.2.dev1/swarmauri/standard/models/concrete/OpenAIToolModel.py

@@ -0,0 +1,44 @@
+from dataclasses import dataclass
+from openai import OpenAI
+from swarmauri.standard.models.base.ModelBase import ModelBase
+
+
+@dataclass
+class OpenAIToolModel(ModelBase):
+    allowed_models = ['gpt-4o', 
+    'gpt-4o-2024-05-13',
+    'gpt-4-turbo', 
+    'gpt-4-turbo-2024-04-09',
+    'gpt-4-turbo-preview',
+    'gpt-4-0125-preview',
+    'gpt-4-1106-preview',
+    'gpt-4',
+    'gpt-4-0613',
+    'gpt-3.5-turbo',
+    'gpt-3.5-turbo-0125',
+    'gpt-3.5-turbo-1106',
+    'gpt-3.5-turbo-0613']
+    api_key: str = ""
+    model_name: str = "gpt-3.5-turbo-0125"
+
+    def __post_init__(self):
+        self._validate_model_name()
+        self.client = OpenAI(api_key=self.api_key)
+
+    def _validate_model_name(self):
+        if self.model_name not in self.allowed_models:
+            raise ValueError(f"Invalid model name: {self.model_name}. Allowed models are: {self.allowed_models}")
+
+
+    def predict(self, messages, tools=None, tool_choice=None, temperature=0.7, max_tokens=1024):
+        if tools and not tool_choice:
+            tool_choice = "auto"
+        response = self.client.chat.completions.create(
+            model=self.model_name,
+            messages=messages,
+            temperature=temperature,
+            max_tokens=max_tokens,
+            tools=tools,
+            tool_choice=tool_choice,
+        )
+        return response

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri/standard/tools/base/ToolBase.py

@@ -3,8 +3,8 @@ from typing import Optional, List, Any
 from dataclasses import dataclass, field, asdict
 import json
 from swarmauri.core.tools.ITool import ITool
-from swarmauri.core.BaseComponent import BaseComponent, ResourceTypes
 from swarmauri.standard.tools.concrete.Parameter import Parameter
+from swarmauri.core.BaseComponent import BaseComponent, ResourceTypes
 
 @dataclass
 class ToolBase(ITool, BaseComponent, ABC):
@@ -21,8 +21,6 @@ class ToolBase(ITool, BaseComponent, ABC):
         if not self.description:
             raise ValueError('Tool must have a description.')
 
-
-
     @property
     def parameters(self):
         return self._parameters

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: swarmauri
-Version: 0.3.0.dev37
+Version: 0.3.2.dev1
 Summary: This repository includes core interfaces, standard ABCs and concrete references, third party plugins, and experimental modules for the swarmaURI framework.
 Home-page: http://github.com/swarmauri/swarmauri-sdk
 Author: Jacob Stewart

{swarmauri-0.3.0.dev37 → swarmauri-0.3.2.dev1}/swarmauri.egg-info/SOURCES.txt

@@ -299,7 +299,6 @@ swarmauri/standard/models/__init__.py
 swarmauri/standard/models/base/ModelBase.py
 swarmauri/standard/models/base/__init__.py
 swarmauri/standard/models/concrete/AnthropicModel.py
-swarmauri/standard/models/concrete/AzureGPT.py
 swarmauri/standard/models/concrete/CohereModel.py
 swarmauri/standard/models/concrete/GeminiProModel.py
 swarmauri/standard/models/concrete/GroqModel.py

swarmauri-0.3.0.dev37/swarmauri/standard/models/base/ModelBase.py

@@ -1,28 +0,0 @@
-from abc import ABC, abstractmethod
-from typing import Any
-from swarmauri.core.models.IModel import IModel
-
-class ModelBase(IModel, ABC):
-    """
-    Concrete implementation of the IModel abstract base class.
-    This version includes managing the model name through a property and a setter.
-    """
-    @abstractmethod
-    def __init__(self, model_name: str):
-        self._model_name = model_name
-    
-    @property
-    def model_name(self):
-        return self._model_name
-    
-    @model_name.setter
-    def model_name(self, value: str) -> None:
-        """
-        Property setter that sets the name of the model.
-
-        Parameters:
-        - value (str): The new name of the model.
-        """
-        self._model_name = value
-       
-    

swarmauri-0.3.0.dev37/swarmauri/standard/models/concrete/AzureGPT.py

@@ -1,67 +0,0 @@
-import json
-from openai import AzureOpenAI
-from ..base.ModelBase import ModelBase
-from ....core.models.IPredict import IPredict
-
-class AzureGPT(ModelBase, IPredict):
-    def __init__(self, azure_endpoint: str, api_key: str, api_version: str, model_name: str):
-        """
-        Initialize the Azure model with an API key.
-
-        Parameters:
-        - api_key (str): Your OpenAI API key.
-        """
-        self.azure_endpoint = azure_endpoint
-        self.api_key = api_key
-        self.api_version = api_version
-        self.client = AzureOpenAI(
-                azure_endpoint = azure_endpoint, 
-                api_key = api_key,  
-                api_version = api_version
-            )
-        super().__init__(model_name)
-       
-
-    
-    def predict(self, messages, temperature=0.7, max_tokens=256, enable_json=True):
-        """
-        Generate predictions using the OpenAI model.
-
-        Parameters:
-        - messages: Input data/messages for the model.
-        - temperature (float): Sampling temperature.
-        - max_tokens (int): Maximum number of tokens to generate.
-        - enable_json (bool): Format response as JSON.
-        
-        Returns:
-        - The generated message content.
-        """
-        
-        if enable_json:
-            response = self.client.chat.completions.create(
-                model=self.model_name,
-                messages=messages,
-                temperature=temperature,
-                response_format={ "type": "json_object" },
-                max_tokens=max_tokens,
-                top_p=1,
-                frequency_penalty=0,
-                presence_penalty=0,
-                stop=None
-            )
-        else:
-            response = self.client.chat.completions.create(
-                model=self.model_name,
-                messages=messages,
-                temperature=temperature,
-                max_tokens=max_tokens,
-                top_p=1,
-                frequency_penalty=0,
-                presence_penalty=0,
-                stop=None
-            )
-        
-        result = response.json()
-        message_content = json.loads(result['choices'][0]['message']['content'])
-        
-        return message_content

swarmauri-0.3.0.dev37/swarmauri/standard/models/concrete/GroqToolModel.py

@@ -1,27 +0,0 @@
-from groq import Groq
-from swarmauri.standard.models.base.ModelBase import ModelBase
-from swarmauri.core.models.IPredict import IPredict
-
-class GroqToolModel(ModelBase, IPredict):
-    allowed_models = ['llama3-8b-8192', 'llama3-70b-8192', 'mixtral-8x7b-32768', 'gemma-7b-it']
-
-    def __init__(self, api_key: str, model_name: str = 'mixtral-8x7b-32768'):
-        if model_name not in self.allowed_models:
-            raise ValueError(f"Model name '{model_name}' is not supported. Choose from {self.allowed_models}")
-        
-        self.client = Groq(api_key=api_key)
-        super().__init__(model_name)
-        
-
-    def predict(self, messages, tools=None, tool_choice=None, temperature=0.7, max_tokens=1024):
-        if tools and not tool_choice:
-            tool_choice = "auto"
-        response = self.client.chat.completions.create(
-            model=self.model_name,
-            messages=messages,
-            temperature=temperature,
-            max_tokens=max_tokens,
-            tools=tools,
-            tool_choice=tool_choice,
-        )
-        return response

swarmauri-0.3.0.dev37/swarmauri/standard/models/concrete/OpenAIToolModel.py

@@ -1,21 +0,0 @@
-from openai import OpenAI
-from swarmauri.standard.models.base.ModelBase import ModelBase
-from swarmauri.core.models.IPredict import IPredict
-
-class OpenAIToolModel(ModelBase, IPredict):
-    def __init__(self, api_key: str, model_name: str = "gpt-3.5-turbo-0125"):
-        self.client = OpenAI(api_key=api_key)
-        super().__init__(model_name)
-
-    def predict(self, messages, tools=None, tool_choice=None, temperature=0.7, max_tokens=1024):
-        if tools and not tool_choice:
-            tool_choice = "auto"
-        response = self.client.chat.completions.create(
-            model=self.model_name,
-            messages=messages,
-            temperature=temperature,
-            max_tokens=max_tokens,
-            tools=tools,
-            tool_choice=tool_choice,
-        )
-        return response