# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
import os
from typing import Any, Dict, List, Optional, Type, Union
from pydantic import BaseModel
from camel.configs import WATSONX_API_PARAMS, WatsonXConfig
from camel.logger import get_logger
from camel.messages import OpenAIMessage
from camel.models import BaseModelBackend
from camel.models._utils import try_modify_message_with_format
from camel.types import ChatCompletion, ModelType
from camel.utils import (
BaseTokenCounter,
OpenAITokenCounter,
api_keys_required,
)
logger = get_logger(__name__)
[docs]
class WatsonXModel(BaseModelBackend):
r"""WatsonX API in a unified BaseModelBackend interface.
Args:
model_type (Union[ModelType, str]): Model type for which a backend is
created, one of WatsonX series.
model_config_dict (Optional[Dict[str, Any]], optional): A dictionary
that will be fed into :obj:`ModelInference.chat()`.
If :obj:`None`, :obj:`WatsonXConfig().as_dict()` will be used.
(default: :obj:`None`)
api_key (Optional[str], optional): The API key for authenticating with
the WatsonX service. (default: :obj:`None`)
url (Optional[str], optional): The url to the WatsonX service.
(default: :obj:`None`)
project_id (Optional[str], optional): The project ID authenticating
with the WatsonX service. (default: :obj:`None`)
token_counter (Optional[BaseTokenCounter], optional): Token counter to
use for the model. If not provided, :obj:`OpenAITokenCounter(
ModelType.GPT_4O_MINI)` will be used.
(default: :obj:`None`)
timeout (Optional[float], optional): The timeout value in seconds for
API calls. If not provided, will fall back to the MODEL_TIMEOUT
environment variable or default to 180 seconds.
(default: :obj:`None`)
"""
@api_keys_required(
[
("api_key", 'WATSONX_API_KEY'),
("project_id", 'WATSONX_PROJECT_ID'),
]
)
def __init__(
self,
model_type: Union[ModelType, str],
model_config_dict: Optional[Dict[str, Any]] = None,
api_key: Optional[str] = None,
url: Optional[str] = None,
project_id: Optional[str] = None,
token_counter: Optional[BaseTokenCounter] = None,
timeout: Optional[float] = None,
):
from ibm_watsonx_ai import APIClient, Credentials
from ibm_watsonx_ai.foundation_models import ModelInference
if model_config_dict is None:
model_config_dict = WatsonXConfig().as_dict()
api_key = api_key or os.environ.get("WATSONX_API_KEY")
url = url or os.environ.get(
"WATSONX_URL", "https://jp-tok.ml.cloud.ibm.com"
)
project_id = project_id or os.environ.get("WATSONX_PROJECT_ID")
timeout = timeout or float(os.environ.get("MODEL_TIMEOUT", 180))
super().__init__(
model_type, model_config_dict, api_key, url, token_counter, timeout
)
self._project_id = project_id
credentials = Credentials(api_key=self._api_key, url=self._url)
client = APIClient(credentials, project_id=self._project_id)
self._model = ModelInference(
model_id=self.model_type,
api_client=client,
params=model_config_dict,
)
def _to_openai_response(self, response: Dict[str, Any]) -> ChatCompletion:
r"""Convert WatsonX response to OpenAI format."""
if not response:
raise ValueError("Empty response from WatsonX API")
# Extract usage information
usage = response.get("usage", {})
# Create OpenAI-compatible response
obj = ChatCompletion.construct(
id=response.get("id", ""),
choices=response.get("choices", []),
created=response.get("created"),
model=self.model_type,
object="chat.completion",
usage=usage,
)
return obj
@property
def token_counter(self) -> BaseTokenCounter:
r"""Initialize the token counter for the model backend.
Returns:
BaseTokenCounter: The token counter following the model's
tokenization style.
"""
if not self._token_counter:
self._token_counter = OpenAITokenCounter(
model=ModelType.GPT_4O_MINI
)
return self._token_counter
def _prepare_request(
self,
messages: List[OpenAIMessage],
response_format: Optional[Type[BaseModel]] = None,
tools: Optional[List[Dict[str, Any]]] = None,
) -> Dict[str, Any]:
import copy
request_config = copy.deepcopy(self.model_config_dict)
if tools:
request_config["tools"] = tools
elif response_format:
try_modify_message_with_format(messages[-1], response_format)
request_config["response_format"] = {"type": "json_object"}
return request_config
def _run(
self,
messages: List[OpenAIMessage],
response_format: Optional[Type[BaseModel]] = None,
tools: Optional[List[Dict[str, Any]]] = None,
) -> ChatCompletion:
r"""Runs inference of WatsonX chat completion.
Args:
messages (List[OpenAIMessage]): Message list with the chat history
in OpenAI API format.
response_format (Optional[Type[BaseModel]], optional): The
response format. (default: :obj:`None`)
tools (Optional[List[Dict[str, Any]]], optional): tools to use.
(default: :obj:`None`)
Returns:
ChatCompletion.
"""
try:
request_config = self._prepare_request(
messages, response_format, tools
)
# WatsonX expects messages as a list of dictionaries
response = self._model.chat(
messages=messages,
params=request_config,
tools=tools,
)
openai_response = self._to_openai_response(response)
return openai_response
except Exception as e:
logger.error(f"Unexpected error when calling WatsonX API: {e!s}")
raise
async def _arun(
self,
messages: List[OpenAIMessage],
response_format: Optional[Type[BaseModel]] = None,
tools: Optional[List[Dict[str, Any]]] = None,
) -> ChatCompletion:
r"""Runs inference of WatsonX chat completion asynchronously.
Args:
messages (List[OpenAIMessage]): Message list with the chat history
in OpenAI API format.
response_format (Optional[Type[BaseModel]], optional): The
response format. (default: :obj:`None`)
tools (Optional[List[Dict[str, Any]]], optional): tools to use.
(default: :obj:`None`)
Returns:
ChatCompletion.
"""
try:
request_config = self._prepare_request(
messages, response_format, tools
)
# WatsonX expects messages as a list of dictionaries
response = await self._model.achat(
messages=messages,
params=request_config,
tools=tools,
)
openai_response = self._to_openai_response(response)
return openai_response
except Exception as e:
logger.error(f"Unexpected error when calling WatsonX API: {e!s}")
raise
[docs]
def check_model_config(self):
r"""Check whether the model configuration contains any unexpected
arguments to WatsonX API.
Raises:
ValueError: If the model configuration dictionary contains any
unexpected arguments to WatsonX API.
"""
for param in self.model_config_dict:
if param not in WATSONX_API_PARAMS:
raise ValueError(
f"Unexpected argument `{param}` is "
"input into WatsonX model backend."
)
@property
def stream(self) -> bool:
r"""Returns whether the model is in stream mode, which sends partial
results each time.
Returns:
bool: Whether the model is in stream mode.
"""
return False
