File size: 5,194 Bytes
b9d9271
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
# ========= 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, Union

from openai import OpenAI, Stream
from openai.types.chat import (
    ChatCompletion,
    ChatCompletionChunk,
)

from camel.configs import NVIDIA_API_PARAMS, NvidiaConfig
from camel.messages import OpenAIMessage
from camel.models import BaseModelBackend
from camel.types import ModelType
from camel.utils import BaseTokenCounter, OpenAITokenCounter, api_keys_required


class NvidiaModel(BaseModelBackend):
    r"""NVIDIA API in a unified BaseModelBackend interface.

    Args:
        model_type (Union[ModelType, str]): Model for which a backend is
            created, one of NVIDIA series.
        model_config_dict (Optional[Dict[str, Any]], optional): A dictionary
            that will be fed into:obj:`openai.ChatCompletion.create()`. If
            :obj:`None`, :obj:`NvidiaConfig().as_dict()` will be used.
            (default: :obj:`None`)
        api_key (Optional[str], optional): The API key for authenticating with
            the NVIDIA service. (default: :obj:`None`)
        url (Optional[str], optional): The url to the NVIDIA service.
            (default: :obj:`None`)
        token_counter (Optional[BaseTokenCounter], optional): Token counter to
            use for the model. If not provided, :obj:`OpenAITokenCounter(
            ModelType.GPT_4)` will be used.
            (default: :obj:`None`)
    """

    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,
        token_counter: Optional[BaseTokenCounter] = None,
    ) -> None:
        if model_config_dict is None:
            model_config_dict = NvidiaConfig().as_dict()
        api_key = api_key or os.environ.get("NVIDIA_API_KEY")
        url = url or os.environ.get(
            "NVIDIA_API_BASE_URL", "https://integrate.api.nvidia.com/v1"
        )
        super().__init__(
            model_type, model_config_dict, api_key, url, token_counter
        )
        self._client = OpenAI(
            timeout=60,
            max_retries=3,
            api_key=self._api_key,
            base_url=self._url,
        )

    @api_keys_required("NVIDIA_API_KEY")
    def run(
        self,
        messages: List[OpenAIMessage],
    ) -> Union[ChatCompletion, Stream[ChatCompletionChunk]]:
        r"""Runs inference of NVIDIA chat completion.

        Args:
            messages (List[OpenAIMessage]): Message list with the chat history
                in OpenAI API format.

        Returns:
            Union[ChatCompletion, Stream[ChatCompletionChunk]]:
                `ChatCompletion` in the non-stream mode, or
                `Stream[ChatCompletionChunk]` in the stream mode.
        """

        # Remove tool-related parameters if no tools are specified
        config = dict(self.model_config_dict)
        if not config.get('tools'):  # None or empty list
            config.pop('tools', None)
            config.pop('tool_choice', None)

        response = self._client.chat.completions.create(
            messages=messages,
            model=self.model_type,
            **config,
        )
        return response

    @property
    def token_counter(self) -> BaseTokenCounter:
        r"""Initialize the token counter for the model backend.

        Returns:
            OpenAITokenCounter: The token counter following the model's
                tokenization style.
        """

        if not self._token_counter:
            self._token_counter = OpenAITokenCounter(ModelType.GPT_4O_MINI)
        return self._token_counter

    def check_model_config(self):
        r"""Check whether the model configuration contains any
        unexpected arguments to NVIDIA API.

        Raises:
            ValueError: If the model configuration dictionary contains any
                unexpected arguments to NVIDIA API.
        """
        for param in self.model_config_dict:
            if param not in NVIDIA_API_PARAMS:
                raise ValueError(
                    f"Unexpected argument `{param}` is "
                    "input into NVIDIA 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 self.model_config_dict.get('stream', False)