From 3fbc9a2240f395076cfc6b1a5b95928eae4fc5eb Mon Sep 17 00:00:00 2001 From: "yuxin.wang" Date: Tue, 6 Feb 2024 14:13:12 +0800 Subject: [PATCH] add moonshot (#30) * add moonshot * Update version --------- Co-authored-by: wangyuxin --- README.md | 3 +- generate/__init__.py | 4 ++ generate/chat_completion/__init__.py | 7 +++ generate/chat_completion/models/__init__.py | 3 + generate/chat_completion/models/azure.py | 4 +- generate/chat_completion/models/moonshot.py | 64 +++++++++++++++++++++ generate/chat_completion/models/openai.py | 15 +++-- generate/platforms/__init__.py | 2 + generate/platforms/moonshot.py | 9 +++ generate/ui.py | 6 +- generate/version.py | 2 +- pyproject.toml | 5 +- 12 files changed, 110 insertions(+), 14 deletions(-) create mode 100644 generate/chat_completion/models/moonshot.py create mode 100644 generate/platforms/moonshot.py diff --git a/README.md b/README.md index 5d3f9e8..3e9fd2e 100644 --- a/README.md +++ b/README.md @@ -15,12 +15,13 @@ Generate Package 允许用户通过统一的 api 访问跨平台的生成式模 * [Minimax](https://api.minimax.chat/document/guides/chat) * [百度智能云](https://cloud.baidu.com/doc/WENXINWORKSHOP/s/clntwmv7t) * [智谱](https://open.bigmodel.cn/dev/api) +* [月之暗面](https://platform.moonshot.cn/docs) * ... ## Features * **多模态**，支持文本生成，多模态文本生成，结构体生成，图像生成，语音生成... -* **跨平台**，完整支持 OpenAI，Azure，Minimax，智谱，文心一言 在内的国内外多家平台 +* **跨平台**，完整支持 OpenAI，Azure，Minimax，智谱，月之暗面，文心一言 在内的国内外多家平台 * **One API**，统一了不同平台的消息格式，推理参数，接口封装，返回解析，让用户无需关心不同平台的差异 * **异步和流式**，提供流式调用，非流式调用，同步调用，异步调用，异步批量调用，适配不同的应用场景 * **自带电池**，提供 UI，输入检查，参数检查，计费，速率控制，*ChatEngine*, *function call* 等功能 diff --git a/generate/__init__.py b/generate/__init__.py index 4b6d09f..3f65171 100644 --- a/generate/__init__.py +++ b/generate/__init__.py @@ -15,6 +15,8 @@ MinimaxChatParameters, MinimaxProChat, MinimaxProChatParameters, + MoonshotChat, + MoonshotParameters, OpenAIChat, OpenAIChatParameters, RemoteChatCompletionModel, @@ -95,6 +97,8 @@ 'DashScopeChatParameters', 'DashScopeMultiModalChat', 'DashScopeMultiModalChatParameters', + 'MoonshotChat', + 'MoonshotParameters', 'OpenAISpeech', 'OpenAISpeechParameters', 'MinimaxSpeech', diff --git a/generate/chat_completion/__init__.py b/generate/chat_completion/__init__.py index 42e0c84..22f2351 100644 --- a/generate/chat_completion/__init__.py +++ b/generate/chat_completion/__init__.py @@ -21,6 +21,8 @@ MinimaxChatParameters, MinimaxProChat, MinimaxProChatParameters, + MoonshotChat, + MoonshotParameters, OpenAIChat, OpenAIChatParameters, WenxinChat, @@ -46,6 +48,7 @@ (BailianChat, BailianChatParameters), (DashScopeChat, DashScopeChatParameters), (DashScopeMultiModalChat, DashScopeMultiModalChatParameters), + (MoonshotChat, MoonshotParameters), ] ChatModelRegistry: dict[str, tuple[Type[ChatCompletionModel], Type[ModelParameters]]] = { @@ -80,6 +83,10 @@ 'BailianChatParameters', 'DashScopeChat', 'DashScopeChatParameters', + 'DashScopeMultiModalChat', + 'DashScopeMultiModalChatParameters', + 'MoonshotChat', + 'MoonshotParameters', 'MessagePrinter', 'SimpleMessagePrinter', 'get_json_schema', diff --git a/generate/chat_completion/models/__init__.py b/generate/chat_completion/models/__init__.py index 673b56c..fbd2d19 100644 --- a/generate/chat_completion/models/__init__.py +++ b/generate/chat_completion/models/__init__.py @@ -13,6 +13,7 @@ from generate.chat_completion.models.hunyuan import HunyuanChat, HunyuanChatParameters from generate.chat_completion.models.minimax import MinimaxChat, MinimaxChatParameters from generate.chat_completion.models.minimax_pro import MinimaxProChat, MinimaxProChatParameters +from generate.chat_completion.models.moonshot import MoonshotChat, MoonshotParameters from generate.chat_completion.models.openai import OpenAIChat, OpenAIChatParameters from generate.chat_completion.models.wenxin import WenxinChat, WenxinChatParameters from generate.chat_completion.models.zhipu import ( @@ -46,4 +47,6 @@ 'DashScopeChatParameters', 'DashScopeMultiModalChat', 'DashScopeMultiModalChatParameters', + 'MoonshotChat', + 'MoonshotParameters', ] diff --git a/generate/chat_completion/models/azure.py b/generate/chat_completion/models/azure.py index 8055592..e844be2 100644 --- a/generate/chat_completion/models/azure.py +++ b/generate/chat_completion/models/azure.py @@ -60,7 +60,7 @@ def generate(self, prompt: Prompt, **kwargs: Unpack[OpenAIChatParametersDict]) - parameters = self.parameters.clone_with_changes(**kwargs) request_parameters = self._get_request_parameters(messages, parameters) response = self.http_client.post(request_parameters) - output = parse_openai_model_reponse(response.json()) + output = parse_openai_model_reponse(response.json(), model_type=self.model_type) output.model_info.type = self.model_type return output @@ -70,7 +70,7 @@ async def async_generate(self, prompt: Prompt, **kwargs: Unpack[OpenAIChatParame parameters = self.parameters.clone_with_changes(**kwargs) request_parameters = self._get_request_parameters(messages, parameters) response = await self.http_client.async_post(request_parameters=request_parameters) - output = parse_openai_model_reponse(response.json()) + output = parse_openai_model_reponse(response.json(), model_type=self.model_type) output.model_info.type = self.model_type return output diff --git a/generate/chat_completion/models/moonshot.py b/generate/chat_completion/models/moonshot.py new file mode 100644 index 0000000..d605959 --- /dev/null +++ b/generate/chat_completion/models/moonshot.py @@ -0,0 +1,64 @@ +from __future__ import annotations + +from typing import AsyncIterator, ClassVar, Iterator, Optional + +from pydantic import PositiveInt +from typing_extensions import Unpack, override + +from generate.chat_completion.message import Prompt +from generate.chat_completion.model_output import ChatCompletionOutput, ChatCompletionStreamOutput +from generate.chat_completion.models.openai import OpenAIChat +from generate.http import HttpClient +from generate.model import ModelParameters, ModelParametersDict +from generate.platforms import MoonshotSettings +from generate.types import Probability, Temperature + + +class MoonshotParameters(ModelParameters): + temperature: Optional[Temperature] = None + top_p: Optional[Probability] = None + max_tokens: Optional[PositiveInt] = None + + +class MoonshotParametersDict(ModelParametersDict, total=False): + temperature: Temperature + top_p: Probability + max_tokens: PositiveInt + + +class MoonshotChat(OpenAIChat): + model_type: ClassVar[str] = 'moonshot' + + parameters: MoonshotParameters + settings: MoonshotSettings + + def __init__( + self, + model: str = 'moonshot-v1-8k', + parameters: MoonshotParameters | None = None, + settings: MoonshotSettings | None = None, + http_client: HttpClient | None = None, + ) -> None: + self.parameters = parameters or MoonshotParameters() + self.settings = settings or MoonshotSettings() # type: ignore + self.http_client = http_client or HttpClient() + self.model = model + + @override + def generate(self, prompt: Prompt, **kwargs: Unpack[MoonshotParametersDict]) -> ChatCompletionOutput: + return super().generate(prompt, **kwargs) + + @override + async def async_generate(self, prompt: Prompt, **kwargs: Unpack[MoonshotParametersDict]) -> ChatCompletionOutput: + return await super().async_generate(prompt, **kwargs) + + @override + def stream_generate(self, prompt: Prompt, **kwargs: Unpack[MoonshotParametersDict]) -> Iterator[ChatCompletionStreamOutput]: + yield from super().stream_generate(prompt, **kwargs) + + @override + async def async_stream_generate( + self, prompt: Prompt, **kwargs: Unpack[MoonshotParametersDict] + ) -> AsyncIterator[ChatCompletionStreamOutput]: + async for i in super().async_stream_generate(prompt, **kwargs): + yield i diff --git a/generate/chat_completion/models/openai.py b/generate/chat_completion/models/openai.py index c3df52c..3a8d809 100644 --- a/generate/chat_completion/models/openai.py +++ b/generate/chat_completion/models/openai.py @@ -214,6 +214,13 @@ def calculate_cost(model_name: str, input_tokens: int, output_tokens: int) -> fl return (0.03 * dollar_to_yuan) * (input_tokens / 1000) + (0.06 * dollar_to_yuan) * (output_tokens / 1000) if 'gpt-3.5-turbo' in model_name: return (0.001 * dollar_to_yuan) * (input_tokens / 1000) + (0.002 * dollar_to_yuan) * (output_tokens / 1000) + if 'moonshot' in model_name: + if '8k' in model_name: + return 0.012 * (input_tokens / 1000) + 0.012 * (output_tokens / 1000) + if '32k' in model_name: + return 0.024 * (input_tokens / 1000) + 0.024 * (output_tokens / 1000) + if '128k' in model_name: + return 0.06 * (input_tokens / 1000) + 0.06 * (output_tokens / 1000) return None @@ -242,7 +249,7 @@ def _convert_to_assistant_message(message: dict[str, Any]) -> AssistantMessage: return AssistantMessage(content=message.get('content') or '', function_call=function_call, tool_calls=tool_calls) -def parse_openai_model_reponse(response: ResponseValue) -> ChatCompletionOutput: +def parse_openai_model_reponse(response: ResponseValue, model_type: str) -> ChatCompletionOutput: message = _convert_to_assistant_message(response['choices'][0]['message']) extra = {'usage': response['usage']} if system_fingerprint := response.get('system_fingerprint'): @@ -253,7 +260,7 @@ def parse_openai_model_reponse(response: ResponseValue) -> ChatCompletionOutput: finish_reason = finish_details['type'] if (finish_details := choice.get('finish_details')) else None return ChatCompletionOutput( - model_info=ModelInfo(task='chat_completion', type='openai', name=response['model']), + model_info=ModelInfo(task='chat_completion', type=model_type, name=response['model']), message=message, finish_reason=finish_reason or '', cost=calculate_cost(response['model'], response['usage']['prompt_tokens'], response['usage']['completion_tokens']), @@ -380,7 +387,7 @@ def generate(self, prompt: Prompt, **kwargs: Unpack[OpenAIChatParametersDict]) - parameters = self.parameters.clone_with_changes(**kwargs) request_parameters = self._get_request_parameters(messages, parameters) response = self.http_client.post(request_parameters) - return parse_openai_model_reponse(response.json()) + return parse_openai_model_reponse(response.json(), model_type=self.model_type) @override async def async_generate(self, prompt: Prompt, **kwargs: Unpack[OpenAIChatParametersDict]) -> ChatCompletionOutput: @@ -388,7 +395,7 @@ async def async_generate(self, prompt: Prompt, **kwargs: Unpack[OpenAIChatParame parameters = self.parameters.clone_with_changes(**kwargs) request_parameters = self._get_request_parameters(messages, parameters) response = await self.http_client.async_post(request_parameters=request_parameters) - return parse_openai_model_reponse(response.json()) + return parse_openai_model_reponse(response.json(), model_type=self.model_type) def _get_stream_request_parameters(self, messages: Messages, parameters: OpenAIChatParameters) -> HttpxPostKwargs: http_parameters = self._get_request_parameters(messages, parameters) diff --git a/generate/platforms/__init__.py b/generate/platforms/__init__.py index 7dff194..3576afe 100644 --- a/generate/platforms/__init__.py +++ b/generate/platforms/__init__.py @@ -5,6 +5,7 @@ from generate.platforms.dashscope import DashScopeSettings from generate.platforms.hunyuan import HunyuanSettings from generate.platforms.minimax import MinimaxSettings +from generate.platforms.moonshot import MoonshotSettings from generate.platforms.openai import OpenAISettings from generate.platforms.zhipu import ZhipuSettings @@ -19,4 +20,5 @@ 'BailianSettings', 'HunyuanSettings', 'DashScopeSettings', + 'MoonshotSettings', ] diff --git a/generate/platforms/moonshot.py b/generate/platforms/moonshot.py new file mode 100644 index 0000000..6523187 --- /dev/null +++ b/generate/platforms/moonshot.py @@ -0,0 +1,9 @@ +from pydantic import SecretStr +from pydantic_settings import BaseSettings, SettingsConfigDict + + +class MoonshotSettings(BaseSettings): + model_config = SettingsConfigDict(extra='ignore', env_prefix='moonshot_', env_file='.env') + + api_key: SecretStr + api_base: str = 'https://api.moonshot.cn/v1' diff --git a/generate/ui.py b/generate/ui.py index 31655ef..1b85868 100644 --- a/generate/ui.py +++ b/generate/ui.py @@ -48,7 +48,7 @@ def chat_model(self) -> ChatCompletionModel: def get_avatars() -> List[Avatar]: avatar_map = { 'openai': 'https://mrvian.com/wp-content/uploads/2023/02/logo-open-ai.png', - 'wenxin': 'https://nlp-eb.cdn.bcebos.com/static/eb/asset/robin.e9dc83e5.png', + 'wenxin': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/wzkPgl.png', 'bailian': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/kHZBrw.png', 'dashscope_multimodal': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/kHZBrw.png', 'dashscope': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/kHZBrw.png', @@ -56,7 +56,8 @@ def get_avatars() -> List[Avatar]: 'minimax': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/lvMJ2T.png', 'minimax_pro': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/lvMJ2T.png', 'zhipu': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/HIntEu.png', - 'baichuan': 'https://ai.tboxn.com/wp-content/uploads/2023/08/%E7%99%BE%E5%B7%9D%E5%A4%A7%E6%A8%A1%E5%9E%8B.png', + 'baichuan': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/fODcq1.png', + 'moonshot': 'https://yuxin-wang.oss-cn-beijing.aliyuncs.com/uPic/hc2Ygt.png', } return [Avatar(name=k, url=v) for k, v in avatar_map.items()] @@ -75,6 +76,7 @@ def get_generate_settings() -> List[Any]: 'wenxin', 'baichuan', 'minimax_pro', + 'moonshot', ], ) model_id = TextInput( diff --git a/generate/version.py b/generate/version.py index e1424ed..73e3bb4 100644 --- a/generate/version.py +++ b/generate/version.py @@ -1 +1 @@ -__version__ = '0.3.1' +__version__ = '0.3.2' diff --git a/pyproject.toml b/pyproject.toml index 954f04c..e1992d9 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -1,6 +1,6 @@ [tool.poetry] name = "generate-core" -version = "0.3.1" +version = "0.3.2" description = "文本生成，图像生成，语音生成" authors = ["wangyuxin "] license = "MIT" @@ -40,7 +40,6 @@ select = [ "T", "PT", "RET", - "PL", "TRY", "PERF", ] @@ -52,10 +51,8 @@ ignore = [ "A003", # shadow builtin "ANN1", # self and cls "ANN401", # Dynamically typed expressions (typing.Any) are disallowed in - "PLR0913", # Too many arguments in function definition "TRY003", # Avoid specifying long messages outside the exception class "PLC0414", # reimport - 'PLR0912', # too many branches ] exclude = ["playground", "api_docs"] target-version = "py38"