dashscope 1.22.1__py3-none-any.whl → 1.22.2__py3-none-any.whl

dashscope/finetune.py

@@ -0,0 +1,175 @@
+from typing import Union
+
+from dashscope.api_entities.dashscope_response import DashScopeAPIResponse
+from dashscope.client.base_api import (CancelMixin, CreateMixin, DeleteMixin,
+                                       GetStatusMixin, ListMixin, LogMixin,
+                                       StreamEventMixin)
+
+
+class FineTune(CreateMixin, CancelMixin, DeleteMixin, ListMixin,
+               GetStatusMixin, StreamEventMixin, LogMixin):
+    SUB_PATH = 'fine-tunes'
+
+    @classmethod
+    def call(cls,
+             model: str,
+             training_file_ids: Union[list, str],
+             validation_file_ids: Union[list, str] = None,
+             mode: str = None,
+             hyper_parameters: dict = {},
+             api_key: str = None,
+             workspace: str = None,
+             **kwargs) -> DashScopeAPIResponse:
+        """Create fine-tune job
+
+        Args:
+            model (str): The model to be fine-tuned
+            training_file_ids (list, str): Ids of the fine-tune training data,
+                which can be pre-uploaded using the File API.
+            validation_file_ids ([list,str], optional): Ids of the fine-tune
+                validating data, which can be pre-uploaded using the File API.
+            mode (str): The fine-tune mode, sft or efficient_sft.
+            hyper_parameters (dict, optional): The fine-tune hyper parameters.
+                Defaults to empty.
+            api_key (str, optional): The api key. Defaults to None.
+            workspace (str): The dashscope workspace id.
+
+        Returns:
+            DashScopeAPIResponse: The request result.
+        """
+        request = {
+            'model': model,
+            'training_file_ids': training_file_ids,
+            'validation_file_ids': validation_file_ids,
+            'hyper_parameters': hyper_parameters if hyper_parameters else {},
+        }
+        if mode is not None:
+            request['training_type'] = mode
+        if 'finetuned_output' in kwargs:
+            request['finetuned_output'] = kwargs['finetuned_output']
+        return super().call(request,
+                            api_key=api_key,
+                            workspace=workspace,
+                            **kwargs)
+
+    @classmethod
+    def cancel(cls,
+               job_id: str,
+               api_key: str = None,
+               workspace: str = None,
+               **kwargs) -> DashScopeAPIResponse:
+        """Cancel a running fine-tune job.
+
+        Args:
+            job_id (str): The fine-tune job id.
+            api_key (str, optional): The api api_key, can be None,
+                if None, will get by default rule(TODO: api key doc).
+            workspace (str): The dashscope workspace id.
+
+        Returns:
+            DashScopeAPIResponse: The request result.
+        """
+        return super().cancel(job_id, api_key, workspace=workspace, **kwargs)
+
+    @classmethod
+    def list(cls,
+             page=1,
+             page_size=10,
+             api_key: str = None,
+             workspace: str = None,
+             **kwargs) -> DashScopeAPIResponse:
+        """List fine-tune job.
+
+        Args:
+            api_key (str, optional): The api key
+            page (int, optional): Page number. Defaults to 1.
+            page_size (int, optional): Items per page. Defaults to 10.
+            workspace (str): The dashscope workspace id.
+
+        Returns:
+            DashScopeAPIResponse: The fine-tune jobs in the result.
+        """
+        return super().list(page,
+                            page_size,
+                            api_key,
+                            workspace=workspace,
+                            **kwargs)
+
+    @classmethod
+    def get(cls,
+            job_id: str,
+            api_key: str = None,
+            workspace: str = None,
+            **kwargs) -> DashScopeAPIResponse:
+        """Get fine-tune job information.
+
+        Args:
+            job_id (str): The fine-tune job id
+            api_key (str, optional): The api key. Defaults to None.
+            workspace (str): The dashscope workspace id.
+
+        Returns:
+            DashScopeAPIResponse: The job info
+        """
+        return super().get(job_id, api_key, workspace=workspace, **kwargs)
+
+    @classmethod
+    def delete(cls,
+               job_id: str,
+               api_key: str = None,
+               workspace: str = None,
+               **kwargs) -> DashScopeAPIResponse:
+        """Delete a fine-tune job.
+
+        Args:
+            job_id (str): The fine-tune job id.
+            api_key (str, optional): The api key. Defaults to None.
+            workspace (str): The dashscope workspace id.
+
+        Returns:
+            DashScopeAPIResponse: The delete result.
+        """
+        return super().delete(job_id, api_key, workspace=workspace, **kwargs)
+
+    @classmethod
+    def stream_events(cls,
+                      job_id: str,
+                      api_key: str = None,
+                      workspace: str = None,
+                      **kwargs) -> DashScopeAPIResponse:
+        """Get fine-tune job events.
+
+        Args:
+            job_id (str): The fine-tune job id
+            api_key (str, optional): the api key. Defaults to None.
+            workspace (str): The dashscope workspace id.
+
+        Returns:
+            DashScopeAPIResponse: The job log events.
+        """
+        return super().stream_events(job_id,
+                                     api_key,
+                                     workspace=workspace,
+                                     **kwargs)
+
+    @classmethod
+    def logs(cls,
+             job_id: str,
+             offset=1,
+             line=1000,
+             api_key: str = None,
+             workspace: str = None,
+             **kwargs) -> DashScopeAPIResponse:
+        """Get log of the job.
+
+        Args:
+            job_id (str): The job id(used for fine-tune)
+            offset (int, optional): start log line. Defaults to 1.
+            line (int, optional): total line return. Defaults to 1000.
+            api_key (str, optional): The api key. Defaults to None.
+            workspace (str): The dashscope workspace id.
+
+        Returns:
+            DashScopeAPIResponse: The response
+        """
+        return super().logs(job_id, offset, line, workspace=workspace)

dashscope/version.py

@@ -1 +1 @@
-__version__ = '1.22.1'
+__version__ = '1.22.2'

{dashscope-1.22.1.dist-info → dashscope-1.22.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dashscope
-Version: 1.22.1
+Version: 1.22.2
 Summary: dashscope client sdk library
 Home-page: https://dashscope.aliyun.com/
 Author: Alibaba Cloud
@@ -18,7 +18,6 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Requires-Python: >=3.8.0
 Description-Content-Type: text/markdown
-License-File: LICENSE
 Requires-Dist: aiohttp
 Requires-Dist: requests
 Requires-Dist: websocket-client
@@ -77,7 +76,7 @@ else:
 
 ## API Key Authentication
 
-The SDK uses API key for authentication. Please refer to [official documentation](https://dashscope.aliyun.com) regarding how to obtain your api-key.
+The SDK uses API key for authentication. Please refer to [official documentation for alibabacloud china](https://www.alibabacloud.com/help/en/model-studio/) and [official documentation for alibabacloud international](https://www.alibabacloud.com/help/en/model-studio/) regarding how to obtain your api-key.
 
 ### Using the API Key
 
@@ -222,3 +221,5 @@ Coming soon.
 
 ## License
 This project is licensed under the Apache License (Version 2.0).
+
+

{dashscope-1.22.1.dist-info → dashscope-1.22.2.dist-info}/RECORD

@@ -1,26 +1,27 @@
 dashscope/__init__.py,sha256=_9PaKXKpYc6PXO35BjcH3R0As9sTsXGOW2iQ6R5HAVc,3011
 dashscope/cli.py,sha256=amegoTkGOs6TlHMdoo4JVOqBePo3lGs745rc7leEyrE,24020
+dashscope/deployment.py,sha256=ljmVi-ny6SjEs8v4oIGNWIw8UQTorE7dl5QJv7dEPIQ,5728
+dashscope/file.py,sha256=Dv2Fz3DLbcye2uuQxyQwRM7ky27OthouLXIpSQagQy4,3324
 dashscope/files.py,sha256=QgJjwhtn9F548nCA8jD8OvE6aQEj-20hZqJgYXsUdQU,3930
+dashscope/finetune.py,sha256=_tflDUvu0KagSoCzLaf0hofpG_P8NU6PylL8CPjVhrA,6243
 dashscope/model.py,sha256=UPOn1qMYFhX-ovXi3BMxZEBk8qOK7WLJOYHMbPZwYBo,1440
 dashscope/models.py,sha256=1-bc-Ue68zurgu_y6RhfFr9uzeQMF5AZq-C32lJGMGU,1224
-dashscope/version.py,sha256=sUm-r83NIqFP93EJcBrI6UFP5e6fmMbnDa6TglGqS1o,23
+dashscope/version.py,sha256=VzBJug2jhAHnqtcPEyOBjewkNpFsBuxDvTGCG_kKqzk,23
 dashscope/aigc/__init__.py,sha256=xmdalVw7wS0cLIuU8Q0qk0q8XGw-iGk8NnQwAQZ3jAc,391
-dashscope/aigc/chat_completion.py,sha256=28puJrtHkqTMZlyKLXb3UjSuICy2OPvysj9a6oNU7Vs,14585
 dashscope/aigc/code_generation.py,sha256=KAJVrGp6tiNFBBg64Ovs9RfcP5SrIhrbW3wdA89NKso,10885
 dashscope/aigc/conversation.py,sha256=xRoJlCR-IXHjSdkDrK74a9ut1FJg0FZhTNXZAJC18MA,14231
 dashscope/aigc/generation.py,sha256=53oMCmN5ZbqeqAsKxmdunXlRh-XP8ZtnA7hB2id4Koo,17897
 dashscope/aigc/image_synthesis.py,sha256=UWHW-nvf7_aDZKr4uZDusVHjqWr9TSZjCsZI8YSWaek,11052
-dashscope/aigc/multimodal_conversation.py,sha256=SlNnnsUPV19gdx8fYJAtsMFWPNGY6vhk5IGHZ5ZczpI,5369
+dashscope/aigc/multimodal_conversation.py,sha256=NPg4gjG24zKdu7lQ4wH2v2m9n_rfZERTLf5tjy5fQyg,5523
 dashscope/aigc/video_synthesis.py,sha256=gbp5XG_DMVN0JnBOgxftdwsCIyrV7nM2gVnlyO8X-ek,10400
 dashscope/api_entities/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dashscope/api_entities/aiohttp_request.py,sha256=ZFbdpJh7SwHnBzbYLhqr_FdcDVRgLVMLhLUS_vXbUGs,10228
 dashscope/api_entities/api_request_data.py,sha256=QHGgIShcQ1jANO1szfsqUGtYix1nD2RN4LyC_fNciCM,5462
 dashscope/api_entities/api_request_factory.py,sha256=iLjOqZkbuQkn9JVduI39XPrhm1UefekAekaNqxFpNLM,5061
 dashscope/api_entities/base_request.py,sha256=cXUL7xqSV8wBr5d-1kx65AO3IsRR9A_ps6Lok-v-MKM,926
-dashscope/api_entities/chat_completion_types.py,sha256=U1nYR5krZYbCedPwlRufJok3hz95mm4kYn2mp20hNAc,12807
-dashscope/api_entities/dashscope_response.py,sha256=FC9zGLNb2iu5fl4bJNcjLnVren_4kNGRW0i9xs4b4Q8,19839
+dashscope/api_entities/dashscope_response.py,sha256=3e-W0UnfPGTsrd_DiZaD_xeq2I5CnrXFMEMRoHHDga8,20131
 dashscope/api_entities/http_request.py,sha256=Pr6mr01uXELK9LwIPXrJAhNtGMkWH3gQYORXShRiQRo,13258
-dashscope/api_entities/websocket_request.py,sha256=XzLj9vsJfBAhYMyRRcbEBoJu_dAVm0HpM60k46lBqBw,16281
+dashscope/api_entities/websocket_request.py,sha256=h3RcvgvXpyzF8xvC4hqNUkev6HGN_YgO5emvqYaXw78,16094
 dashscope/app/__init__.py,sha256=UiN_9i--z84Dw5wUehOh_Tkk_9Gq_td_Kbz1dobBEKg,62
 dashscope/app/application.py,sha256=Cnd62LFpG70XJUo4Oibry9KzXPhPNmNkKFD4R5YuGTA,9343
 dashscope/app/application_response.py,sha256=0pulI3O3z4R4h_YaDwzVimamo3XwTXGy5TiHCzysTBg,7011
@@ -32,6 +33,7 @@ dashscope/audio/__init__.py,sha256=-ZRxrK-gV4QsUtlThIT-XwqB6vmyEsnhxIxdLmhCUuc,6
 dashscope/audio/asr/__init__.py,sha256=VaWX5DRWcB81_5z2o7IPwz6Jrs9vFFJ5GEVarzVOvPY,1004
 dashscope/audio/asr/asr_phrase_manager.py,sha256=EjtbI3zz9UQGS1qv6Yb4zzEMj4OJJVXmwkqZyIrzvEA,7642
 dashscope/audio/asr/recognition.py,sha256=7ApnKJKVp1JS6e6KBZzM8y1yiMB2fEun07S6BOvqfrs,20699
+dashscope/audio/asr/transcribe.py,sha256=HfZYpvpVfvGRAIIIzX65Af33E6vsIFGd_qqhQ8LaNcM,9651
 dashscope/audio/asr/transcription.py,sha256=D8CW0XDqJuEJVmNFJ6qczTysSV3Sz_rzk2C6NIKTtVc,9042
 dashscope/audio/asr/translation_recognizer.py,sha256=j1oKgggSA2HDdEJAOJ4jY78_bK0FGwpfIf_AxbrAzWM,39629
 dashscope/audio/asr/vocabulary.py,sha256=2MHxeaL0ANWk-TILrHhArKSdj0d5M_YHw0cnjB-E4dY,6476
@@ -58,7 +60,7 @@ dashscope/customize/finetunes.py,sha256=iFnEUMGY6FGogVClIJMEeFhUwfYi7gD05Iq1-eMP
 dashscope/embeddings/__init__.py,sha256=-dxHaoxZZVuP-wAGUIa3sNNh8CQwaeWj2UlqsDy1sV4,240
 dashscope/embeddings/batch_text_embedding.py,sha256=P32LFO9v7ehdJsl0c32In94hUET6K6AaGJ_pDRtFqco,8791
 dashscope/embeddings/batch_text_embedding_response.py,sha256=hf17wfhJKvX3ggnigOiTqD4O1lFhx_NG_O1ioKCx0gk,1999
-dashscope/embeddings/multimodal_embedding.py,sha256=yojtGNoT2N54g0jcAYUwNIiwzueun1ouqS0S0tvnyQc,3966
+dashscope/embeddings/multimodal_embedding.py,sha256=EvXoVRExkenpeRE8Nqfas0Vrl5wzCZCJIxfgbOI7Qv0,4058
 dashscope/embeddings/text_embedding.py,sha256=I3zRvuT2HWcaZYH-zrtGcAQmzuLQxFswIHJMiQXfaJQ,2009
 dashscope/io/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dashscope/io/input_output.py,sha256=iZ1X1x1btdoZK2VeC9JsKkag2eaXwqfNT3Q6SrmRi2w,3941
@@ -85,10 +87,9 @@ dashscope/tokenizers/tokenizer.py,sha256=y6P91qTCYo__pEx_0VHAcj9YECfbUdRqZU1fdGT
 dashscope/tokenizers/tokenizer_base.py,sha256=REDhzRyDT13iequ61-a6_KcTy0GFKlihQve5HkyoyRs,656
 dashscope/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dashscope/utils/oss_utils.py,sha256=TlqaMAmVRtBJIm5aIaXsrRZGKc_7cwWQ7liMB2f9Css,7331
-dashscope/utils/temporary_storage.py,sha256=CWBpEmuIthYg5DhU_qu63DbF2dsmqnLdKFkbWF8yX6A,6072
-dashscope-1.22.1.dist-info/LICENSE,sha256=Izp5L1DF1Mbza6qojkqNNWlE_mYLnr4rmzx2EBF8YFw,11413
-dashscope-1.22.1.dist-info/METADATA,sha256=xk089kHugSTlkhuy17skH7fOIT60VivJMcsYUUKAI-4,6661
-dashscope-1.22.1.dist-info/WHEEL,sha256=G16H4A3IeoQmnOrYV4ueZGKSjhipXx8zc8nu9FGlvMA,92
-dashscope-1.22.1.dist-info/entry_points.txt,sha256=e9C3sOf9zDYL0O5ROEGX6FT8w-QK_kaGRWmPZDHAFys,49
-dashscope-1.22.1.dist-info/top_level.txt,sha256=woqavFJK9zas5xTqynmALqOtlafghjsk63Xk86powTU,10
-dashscope-1.22.1.dist-info/RECORD,,
+dashscope-1.22.2.dist-info/LICENSE,sha256=Izp5L1DF1Mbza6qojkqNNWlE_mYLnr4rmzx2EBF8YFw,11413
+dashscope-1.22.2.dist-info/METADATA,sha256=Cf_GT_W-HPppanGcy9RBKHS40yWSfL93_IXwc0SxAzc,6798
+dashscope-1.22.2.dist-info/WHEEL,sha256=G16H4A3IeoQmnOrYV4ueZGKSjhipXx8zc8nu9FGlvMA,92
+dashscope-1.22.2.dist-info/entry_points.txt,sha256=raEp5dOuj8whJ7yqZlDM8WQ5p2RfnGrGNo0QLQEnatY,50
+dashscope-1.22.2.dist-info/top_level.txt,sha256=woqavFJK9zas5xTqynmALqOtlafghjsk63Xk86powTU,10
+dashscope-1.22.2.dist-info/RECORD,,

{dashscope-1.22.1.dist-info → dashscope-1.22.2.dist-info}/entry_points.txt

@@ -1,2 +1,3 @@
 [console_scripts]
 dashscope = dashscope.cli:main
+

dashscope/aigc/chat_completion.py

@@ -1,271 +0,0 @@
-import json
-from typing import Any, Dict, Generator, List, Union
-
-import dashscope
-from dashscope.api_entities.dashscope_response import (GenerationResponse,
-                                                       Message)
-from dashscope.client.base_api import BaseAioApi, CreateMixin
-from dashscope.common import constants
-from dashscope.common.error import InputRequired, ModelRequired
-from dashscope.common.utils import _get_task_group_and_task
-from dashscope.api_entities.chat_completion_types import ChatCompletion, ChatCompletionChunk
-
-
-class Completions(CreateMixin):
-    """Support openai compatible chat completion interface.
-
-    """
-    SUB_PATH = ''
-    @classmethod
-    def create(
-        cls,
-        *,
-        model: str,
-        messages: List[Message],
-        stream: bool = False,
-        temperature: float = None,
-        top_p: float = None,
-        top_k: int = None,
-        stop: Union[List[str], List[List[int]]] = None,
-        max_tokens: int = None,
-        repetition_penalty: float = None,
-        api_key: str = None,
-        workspace: str = None,
-        extra_headers: Dict = None,
-        extra_body: Dict = None,
-        **kwargs
-    ) -> Union[ChatCompletion, Generator[ChatCompletionChunk, None, None]]:
-        """Call openai compatible chat completion model service.
-
-        Args:
-            model (str): The requested model, such as qwen-long
-            messages (list): The generation messages.
-                examples:
-                    [{'role': 'user',
-                      'content': 'The weather is fine today.'},
-                      {'role': 'assistant', 'content': 'Suitable for outings'}]
-            stream(bool, `optional`): Enable server-sent events
-                (ref: https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events)  # noqa E501
-                the result will back partially[qwen-turbo,bailian-v1].
-            temperature(float, `optional`): Used to control the degree
-                of randomness and diversity. Specifically, the temperature
-                value controls the degree to which the probability distribution
-                of each candidate word is smoothed when generating text.
-                A higher temperature value will reduce the peak value of
-                the probability, allowing more low-probability words to be
-                selected, and the generated results will be more diverse;
-                while a lower temperature value will enhance the peak value
-                of the probability, making it easier for high-probability
-                words to be selected, the generated results are more
-                deterministic.
-            top_p(float, `optional`): A sampling strategy, called nucleus
-                sampling, where the model considers the results of the
-                tokens with top_p probability mass. So 0.1 means only
-                the tokens comprising the top 10% probability mass are
-                considered.
-            top_k(int, `optional`): The size of the sample candidate set when generated.  # noqa E501
-                For example, when the value is 50, only the 50 highest-scoring tokens  # noqa E501
-                in a single generation form a randomly sampled candidate set. # noqa E501
-                The larger the value, the higher the randomness generated;  # noqa E501
-                the smaller the value, the higher the certainty generated. # noqa E501
-                The default value is 0, which means the top_k policy is  # noqa E501
-                not enabled. At this time, only the top_p policy takes effect. # noqa E501
-            stop(list[str] or list[list[int]], `optional`): Used to control the generation to stop  # noqa E501
-                when encountering setting str or token ids, the result will not include # noqa E501
-                stop words or tokens.
-            max_tokens(int, `optional`): The maximum token num expected to be output. It should be # noqa E501
-                noted that the length generated by the model will only be less than max_tokens,  # noqa E501
-                not necessarily equal to it. If max_tokens is set too large, the service will # noqa E501
-                directly prompt that the length exceeds the limit. It is generally # noqa E501
-                not recommended to set this value.
-            repetition_penalty(float, `optional`): Used to control the repeatability when generating models.  # noqa E501
-                Increasing repetition_penalty can reduce the duplication of model generation.  # noqa E501
-                1.0 means no punishment.
-            api_key (str, optional): The api api_key, can be None,
-                if None, will get by default rule.
-            workspace (str, optional): The bailian workspace id.
-            **kwargs:
-                timeout: set request timeout.
-        Raises:
-            InvalidInput: The history and auto_history are mutually exclusive.
-
-        Returns:
-            Union[ChatCompletion,
-                  Generator[ChatCompletionChunk, None, None]]: If
-            stream is True, return Generator, otherwise ChatCompletion.
-        """
-        if messages is None or not messages:
-            raise InputRequired('Messages is required!')
-        if model is None or not model:
-            raise ModelRequired('Model is required!')
-        data = {}
-        data['model'] = model
-        data['messages'] = messages
-        if temperature is not None:
-            data['temperature'] = temperature
-        if top_p is not None:
-            data['top_p'] = top_p
-        if top_k is not None:
-            data['top_k'] = top_k
-        if stop is not None:
-            data['stop'] = stop
-        if max_tokens is not None:
-            data[max_tokens] = max_tokens
-        if repetition_penalty is not None:
-            data['repetition_penalty'] = repetition_penalty
-        if extra_body is not None and extra_body:
-            data = {**data, **extra_body}
-        
-        if extra_headers is not None and extra_headers:
-            kwargs = {'headers': extra_headers} if kwargs else {**kwargs, **{'headers': extra_headers}}
-
-        response = super().call(data=data,
-                                path='chat/completions',
-                                base_address=dashscope.base_compatible_api_url,
-                                api_key=api_key,
-                                flattened_output=True,
-                                stream=stream,
-                                workspace=workspace,
-                                **kwargs)
-        if stream:
-            return (ChatCompletionChunk(**item) for _, item in response)
-        else:
-            return ChatCompletion(**response)
-
-
-class AioGeneration(BaseAioApi):
-    task = 'text-generation'
-    """API for AI-Generated Content(AIGC) models.
-
-    """
-    class Models:
-        """@deprecated, use qwen_turbo instead"""
-        qwen_v1 = 'qwen-v1'
-        """@deprecated, use qwen_plus instead"""
-        qwen_plus_v1 = 'qwen-plus-v1'
-
-        bailian_v1 = 'bailian-v1'
-        dolly_12b_v2 = 'dolly-12b-v2'
-        qwen_turbo = 'qwen-turbo'
-        qwen_plus = 'qwen-plus'
-        qwen_max = 'qwen-max'
-
-    @classmethod
-    async def call(
-        cls,
-        model: str,
-        prompt: Any = None,
-        history: list = None,
-        api_key: str = None,
-        messages: List[Message] = None,
-        plugins: Union[str, Dict[str, Any]] = None,
-        workspace: str = None,
-        **kwargs
-    ) -> Union[GenerationResponse, Generator[GenerationResponse, None, None]]:
-        """Call generation model service.
-
-        Args:
-            model (str): The requested model, such as qwen-turbo
-            prompt (Any): The input prompt.
-            history (list):The user provided history, deprecated
-                examples:
-                    [{'user':'The weather is fine today.',
-                    'bot': 'Suitable for outings'}].
-                Defaults to None.
-            api_key (str, optional): The api api_key, can be None,
-                if None, will get by default rule(TODO: api key doc).
-            messages (list): The generation messages.
-                examples:
-                    [{'role': 'user',
-                      'content': 'The weather is fine today.'},
-                      {'role': 'assistant', 'content': 'Suitable for outings'}]
-            plugins (Any): The plugin config. Can be plugins config str, or dict.
-            **kwargs:
-                stream(bool, `optional`): Enable server-sent events
-                    (ref: https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events)  # noqa E501
-                    the result will back partially[qwen-turbo,bailian-v1].
-                temperature(float, `optional`): Used to control the degree
-                    of randomness and diversity. Specifically, the temperature
-                    value controls the degree to which the probability distribution
-                    of each candidate word is smoothed when generating text.
-                    A higher temperature value will reduce the peak value of
-                    the probability, allowing more low-probability words to be
-                    selected, and the generated results will be more diverse;
-                    while a lower temperature value will enhance the peak value
-                    of the probability, making it easier for high-probability
-                    words to be selected, the generated results are more
-                    deterministic, range(0, 2) .[qwen-turbo,qwen-plus].
-                top_p(float, `optional`): A sampling strategy, called nucleus
-                    sampling, where the model considers the results of the
-                    tokens with top_p probability mass. So 0.1 means only
-                    the tokens comprising the top 10% probability mass are
-                    considered[qwen-turbo,bailian-v1].
-                top_k(int, `optional`): The size of the sample candidate set when generated.  # noqa E501
-                    For example, when the value is 50, only the 50 highest-scoring tokens  # noqa E501
-                    in a single generation form a randomly sampled candidate set. # noqa E501
-                    The larger the value, the higher the randomness generated;  # noqa E501
-                    the smaller the value, the higher the certainty generated. # noqa E501
-                    The default value is 0, which means the top_k policy is  # noqa E501
-                    not enabled. At this time, only the top_p policy takes effect. # noqa E501
-                enable_search(bool, `optional`): Whether to enable web search(quark).  # noqa E501
-                    Currently works best only on the first round of conversation.
-                    Default to False, support model: [qwen-turbo].
-                customized_model_id(str, required) The enterprise-specific
-                    large model id, which needs to be generated from the
-                    operation background of the enterprise-specific
-                    large model product, support model: [bailian-v1].
-                result_format(str, `optional`): [message|text] Set result result format. # noqa E501
-                    Default result is text
-                incremental_output(bool, `optional`): Used to control the streaming output mode. # noqa E501
-                    If true, the subsequent output will include the previously input content. # noqa E501
-                    Otherwise, the subsequent output will not include the previously output # noqa E501
-                    content. Default false.
-                stop(list[str] or list[list[int]], `optional`): Used to control the generation to stop  # noqa E501
-                    when encountering setting str or token ids, the result will not include # noqa E501
-                    stop words or tokens.
-                max_tokens(int, `optional`): The maximum token num expected to be output. It should be # noqa E501
-                    noted that the length generated by the model will only be less than max_tokens,  # noqa E501
-                    not necessarily equal to it. If max_tokens is set too large, the service will # noqa E501
-                    directly prompt that the length exceeds the limit. It is generally # noqa E501
-                    not recommended to set this value.
-                repetition_penalty(float, `optional`): Used to control the repeatability when generating models.  # noqa E501
-                    Increasing repetition_penalty can reduce the duplication of model generation.  # noqa E501
-                    1.0 means no punishment.
-            workspace (str): The dashscope workspace id.
-        Raises:
-            InvalidInput: The history and auto_history are mutually exclusive.
-
-        Returns:
-            Union[GenerationResponse,
-                  Generator[GenerationResponse, None, None]]: If
-            stream is True, return Generator, otherwise GenerationResponse.
-        """
-        if (prompt is None or not prompt) and (messages is None
-                                               or not messages):
-            raise InputRequired('prompt or messages is required!')
-        if model is None or not model:
-            raise ModelRequired('Model is required!')
-        task_group, function = _get_task_group_and_task(__name__)
-        if plugins is not None:
-            headers = kwargs.pop('headers', {})
-            if isinstance(plugins, str):
-                headers['X-DashScope-Plugin'] = plugins
-            else:
-                headers['X-DashScope-Plugin'] = json.dumps(plugins)
-            kwargs['headers'] = headers
-        input, parameters = Generation._build_input_parameters(
-            model, prompt, history, messages, **kwargs)
-        response = await super().call(model=model,
-                                      task_group=task_group,
-                                      task=Generation.task,
-                                      function=function,
-                                      api_key=api_key,
-                                      input=input,
-                                      workspace=workspace,
-                                      **parameters)
-        is_stream = kwargs.get('stream', False)
-        if is_stream:
-            return (GenerationResponse.from_api_response(rsp)
-                    async for rsp in response)
-        else:
-            return GenerationResponse.from_api_response(response)