Merge 'integration_2024-08-08_369120306946' into 'master'

merge branch integration_2024-08-08_369120306946 into master

See merge request: !361

Author: liuweiwei-xyz

meta.json

@@ -1,4 +1,4 @@
 {
-	"lasted": "1.0.94",
-	"meta_commit": "92cfd1814f057ab80f1ec26a43284b1038a2771f"
+	"lasted": "1.0.95",
+	"meta_commit": "b75db8c6be327ebe7c1c1a161814a0d99c097ba1"
 }

setup.py

@@ -3,7 +3,7 @@
 from setuptools import setup, find_packages  # noqa: H301
 
 NAME = "volcengine-python-sdk"
-VERSION = "1.0.94"
+VERSION = "1.0.95"
 # To install the library, run the following
 #
 # python setup.py install

volcenginesdkarkruntime/_client.py

@@ -34,6 +34,7 @@ class Ark(SyncAPIClient):
     chat: resources.Chat
     bot_chat: resources.BotChat
     embeddings: resources.Embeddings
+    tokenization: resources.Tokenization
 
     def __init__(
         self,
@@ -88,7 +89,7 @@ def __init__(
         self.chat = resources.Chat(self)
         self.bot_chat = resources.BotChat(self)
         self.embeddings = resources.Embeddings(self)
-        # self.tokenization = resources.Tokenization(self)
+        self.tokenization = resources.Tokenization(self)
         # self.classification = resources.Classification(self)
 
     def _get_endpoint_sts_token(self, endpoint_id: str):
@@ -115,6 +116,7 @@ class AsyncArk(AsyncAPIClient):
     chat: resources.AsyncChat
     bot_chat: resources.AsyncBotChat
     embeddings: resources.AsyncEmbeddings
+    tokenization: resources.AsyncTokenization
 
     def __init__(
         self,
@@ -168,7 +170,7 @@ def __init__(
         self.chat = resources.AsyncChat(self)
         self.bot_chat = resources.AsyncBotChat(self)
         self.embeddings = resources.AsyncEmbeddings(self)
-        # self.tokenization = resources.AsyncTokenization(self)
+        self.tokenization = resources.AsyncTokenization(self)
         # self.classification = resources.AsyncClassification(self)
 
     def _get_endpoint_sts_token(self, endpoint_id: str):

volcenginesdkarkruntime/_constants.py

@@ -2,15 +2,15 @@
 
 import httpx
 
-VERSION = ""
+VERSION = "1.0.0"
 BASE_URL = "https://ark.cn-beijing.volces.com/api/v3"
 
 RAW_RESPONSE_HEADER = "X-Stainless-Raw-Response"
 CLIENT_REQUEST_HEADER = "X-Client-Request-Id"
 SERVER_REQUEST_HEADER = "X-Request-Id"
 
 # default timeout is 1 minutes
-DEFAULT_TIMEOUT = httpx.Timeout(timeout=60.0, connect=60.0)
+DEFAULT_TIMEOUT = httpx.Timeout(timeout=600.0, connect=60.0)
 DEFAULT_MAX_RETRIES = 2
 DEFAULT_CONNECTION_LIMITS = httpx.Limits(
     max_connections=1000, max_keepalive_connections=100

volcenginesdkarkruntime/_utils/_utils.py

@@ -76,7 +76,7 @@ async def wrapper(*args, **kwargs):
 
 def _insert_sts_token(args, kwargs):
     assert len(args) > 0
-    assert "model" in kwargs
+    assert "model" in kwargs, "you need to support model"
 
     ark_client = args[0]._client
     model = kwargs.get("model", "")

volcenginesdkarkruntime/resources/chat/completions.py

@@ -25,6 +25,7 @@
     completion_create_params,
     ChatCompletionStreamOptionsParam,
     ChatCompletionToolParam,
+    ChatCompletionToolChoiceOptionParam
 )
 
 __all__ = ["Completions", "AsyncCompletions"]
@@ -47,7 +48,6 @@ def create(
         model: str,
         frequency_penalty: Optional[float] | None = None,
         function_call: completion_create_params.FunctionCall | None = None,
-        functions: Iterable[completion_create_params.Function] | None = None,
         logit_bias: Optional[Dict[str, int]] | None = None,
         logprobs: Optional[bool] | None = None,
         max_tokens: Optional[int] | None = None,
@@ -59,6 +59,10 @@ def create(
         tools: Iterable[ChatCompletionToolParam] | None = None,
         top_logprobs: Optional[int] | None = None,
         top_p: Optional[float] | None = None,
+        repetition_penalty: Optional[float] | None = None,
+        n: Optional[int] | None = None,
+        tool_choice: ChatCompletionToolChoiceOptionParam | None = None,
+        response_format: completion_create_params.ResponseFormat | None = None,
         user: str | None = None,
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
@@ -72,7 +76,6 @@ def create(
                 "model": model,
                 "frequency_penalty": frequency_penalty,
                 "function_call": function_call,
-                "functions": functions,
                 "logit_bias": logit_bias,
                 "logprobs": logprobs,
                 "max_tokens": max_tokens,
@@ -85,6 +88,10 @@ def create(
                 "top_logprobs": top_logprobs,
                 "top_p": top_p,
                 "user": user,
+                "repetition_penalty": repetition_penalty,
+                "n": n,
+                "tool_choice": tool_choice,
+                "response_format": response_format,
             },
             options=make_request_options(
                 extra_headers=extra_headers,
@@ -115,7 +122,6 @@ async def create(
         model: str,
         frequency_penalty: Optional[float] | None = None,
         function_call: completion_create_params.FunctionCall | None = None,
-        functions: Iterable[completion_create_params.Function] | None = None,
         logit_bias: Optional[Dict[str, int]] | None = None,
         logprobs: Optional[bool] | None = None,
         max_tokens: Optional[int] | None = None,
@@ -128,6 +134,10 @@ async def create(
         top_logprobs: Optional[int] | None = None,
         top_p: Optional[float] | None = None,
         user: str | None = None,
+        repetition_penalty: Optional[float] | None = None,
+        n: Optional[int] | None = None,
+        tool_choice: ChatCompletionToolChoiceOptionParam | None = None,
+        response_format: completion_create_params.ResponseFormat | None = None,
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
@@ -140,7 +150,6 @@ async def create(
                 "model": model,
                 "frequency_penalty": frequency_penalty,
                 "function_call": function_call,
-                "functions": functions,
                 "logit_bias": logit_bias,
                 "logprobs": logprobs,
                 "max_tokens": max_tokens,
@@ -153,6 +162,10 @@ async def create(
                 "top_logprobs": top_logprobs,
                 "top_p": top_p,
                 "user": user,
+                "repetition_penalty": repetition_penalty,
+                "n": n,
+                "tool_choice": tool_choice,
+                "response_format": response_format,
             },
             options=make_request_options(
                 extra_headers=extra_headers,

volcenginesdkarkruntime/resources/tokenization.py

@@ -27,9 +27,8 @@ def with_raw_response(self) -> TokenizationWithRawResponse:
     def create(
         self,
         *,
-        input: Union[str, List[str]],
+        text: Union[str, List[str]],
         model: str,
-        encoding_format: Literal["float", "base64"] = "float",
         user: str | None = None,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -41,10 +40,9 @@ def create(
         return self._post(
             "/tokenization",
             body={
-                "input": input,
+                "text": text,
                 "model": model,
                 "user": user,
-                "encoding_format": encoding_format,
             },
             options=make_request_options(
                 extra_headers=extra_headers,
@@ -65,9 +63,8 @@ def with_raw_response(self) -> AsyncTokenizationWithRawResponse:
     async def create(
         self,
         *,
-        input: Union[str, List[str]],
+        text: Union[str, List[str]],
         model: str,
-        encoding_format: Literal["float", "base64"] = "float",
         user: str | None = None,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -79,10 +76,9 @@ async def create(
         return await self._post(
             "/tokenization",
             body={
-                "input": input,
+                "text": text,
                 "model": model,
                 "user": user,
-                "encoding_format": encoding_format,
             },
             options=make_request_options(
                 extra_headers=extra_headers,

volcenginesdkarkruntime/types/chat/chat_completion_tool_choice_option_param.py

@@ -8,5 +8,5 @@
 __all__ = ["ChatCompletionToolChoiceOptionParam"]
 
 ChatCompletionToolChoiceOptionParam = Union[
-    Literal["none", "auto"], ChatCompletionNamedToolChoiceParam
+    Literal["none", "auto", "required"], ChatCompletionNamedToolChoiceParam
 ]

volcenginesdkarkruntime/types/chat/completion_create_params.py

@@ -209,6 +209,13 @@ class ResponseFormat(TypedDict, total=False):
     type: Literal["text", "json_object"]
     """Must be one of `text` or `json_object`."""
 
+    schema: Optional[Dict[str, object]]
+    """If the request only specifies type=`json_object` and no schema is specified, refer to the openai behavior, 
+    the model outputs an arbitrary json object (depending on the user's instruction in the user prompt/system prompt)
+    
+    Even if the schema is specified, still need to specify the expected json format in user prompt/system prompt
+    """
+
 
 class CompletionCreateParamsNonStreaming(CompletionCreateParamsBase):
     stream: Optional[Literal[False]]

volcenginesdkarkruntime/types/create_embedding_response.py

@@ -16,6 +16,12 @@ class Usage(BaseModel):
 
 
 class CreateEmbeddingResponse(BaseModel):
+    id: str
+    """A unique identifier for the embeddings."""
+
+    created: int
+    """The Unix timestamp (in seconds) of when the embeddings was created."""
+
     data: List[Embedding]
     """The list of embeddings generated by the model."""