vectorvein 0.2.7__py3-none-any.whl → 0.2.9__py3-none-any.whl

vectorvein/chat_clients/openai_compatible_client.py

@@ -1,12 +1,21 @@
 # @Author: Bi Ying
 # @Date:   2024-07-26 14:48:55
+import re
 import json
 from functools import cached_property
-from typing import overload, Generator, AsyncGenerator, Any, Literal, Iterable
-import re
+from typing import overload, Generator, AsyncGenerator, Any, Literal, Iterable, Optional, Dict, List, Union
 
 import httpx
 from openai import OpenAI, AsyncOpenAI, AzureOpenAI, AsyncAzureOpenAI
+from openai._types import Headers, Query, Body
+from openai.types.shared_params.metadata import Metadata
+from openai.types.chat.completion_create_params import ResponseFormat
+from openai.types.chat.chat_completion_modality import ChatCompletionModality
+from openai.types.chat.chat_completion_audio_param import ChatCompletionAudioParam
+from openai.types.chat.chat_completion_reasoning_effort import ChatCompletionReasoningEffort
+from openai.types.chat.chat_completion_stream_options_param import ChatCompletionStreamOptionsParam
+from openai.types.chat.chat_completion_prediction_content_param import ChatCompletionPredictionContentParam
+from anthropic.types.thinking_config_param import ThinkingConfigParam
 
 from .base_client import BaseChatClient, BaseAsyncChatClient
 from .utils import (
@@ -20,6 +29,7 @@ from ..types.enums import ContextLengthControlType, BackendType
 from ..types.llm_parameters import (
     NotGiven,
     NOT_GIVEN,
+    OPENAI_NOT_GIVEN,
     ToolParam,
     ToolChoice,
     OpenAINotGiven,
@@ -27,7 +37,6 @@ from ..types.llm_parameters import (
     Usage,
     ChatCompletionMessage,
     ChatCompletionDeltaMessage,
-    ChatCompletionStreamOptionsParam,
 )
 
 
@@ -93,11 +102,33 @@ class OpenAICompatibleChatClient(BaseChatClient):
         max_tokens: int | None = None,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
-        response_format: dict | None = None,
+        response_format: ResponseFormat | NotGiven = NOT_GIVEN,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
         skip_cutoff: bool = False,
-        **kwargs,
+        audio: Optional[ChatCompletionAudioParam] | OpenAINotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | OpenAINotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        metadata: Optional[Metadata] | OpenAINotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | OpenAINotGiven = NOT_GIVEN,
+        n: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | OpenAINotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | OpenAINotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ChatCompletionReasoningEffort] | OpenAINotGiven = NOT_GIVEN,
+        thinking: ThinkingConfigParam | None | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | OpenAINotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | OpenAINotGiven = NOT_GIVEN,
+        store: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        user: str | OpenAINotGiven = NOT_GIVEN,
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | OpenAINotGiven = NOT_GIVEN,
     ) -> ChatCompletionMessage:
         pass
 
@@ -112,11 +143,33 @@ class OpenAICompatibleChatClient(BaseChatClient):
         max_tokens: int | None = None,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
-        response_format: dict | None = None,
+        response_format: ResponseFormat | NotGiven = NOT_GIVEN,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
         skip_cutoff: bool = False,
-        **kwargs,
+        audio: Optional[ChatCompletionAudioParam] | OpenAINotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | OpenAINotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        metadata: Optional[Metadata] | OpenAINotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | OpenAINotGiven = NOT_GIVEN,
+        n: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | OpenAINotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | OpenAINotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ChatCompletionReasoningEffort] | OpenAINotGiven = NOT_GIVEN,
+        thinking: ThinkingConfigParam | None | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | OpenAINotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | OpenAINotGiven = NOT_GIVEN,
+        store: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        user: str | OpenAINotGiven = NOT_GIVEN,
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | OpenAINotGiven = NOT_GIVEN,
     ) -> Generator[ChatCompletionDeltaMessage, None, None]:
         pass
 
@@ -131,11 +184,33 @@ class OpenAICompatibleChatClient(BaseChatClient):
         max_tokens: int | None = None,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
-        response_format: dict | None = None,
+        response_format: ResponseFormat | NotGiven = NOT_GIVEN,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
         skip_cutoff: bool = False,
-        **kwargs,
+        audio: Optional[ChatCompletionAudioParam] | OpenAINotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | OpenAINotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        metadata: Optional[Metadata] | OpenAINotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | OpenAINotGiven = NOT_GIVEN,
+        n: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | OpenAINotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | OpenAINotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ChatCompletionReasoningEffort] | OpenAINotGiven = NOT_GIVEN,
+        thinking: ThinkingConfigParam | None | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | OpenAINotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | OpenAINotGiven = NOT_GIVEN,
+        store: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        user: str | OpenAINotGiven = NOT_GIVEN,
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | OpenAINotGiven = NOT_GIVEN,
     ) -> ChatCompletionMessage | Generator[ChatCompletionDeltaMessage, Any, None]:
         pass
 
@@ -149,11 +224,33 @@ class OpenAICompatibleChatClient(BaseChatClient):
         max_tokens: int | None = None,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
-        response_format: dict | None = None,
+        response_format: ResponseFormat | NotGiven = NOT_GIVEN,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
         skip_cutoff: bool = False,
-        **kwargs,
+        audio: Optional[ChatCompletionAudioParam] | OpenAINotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | OpenAINotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        metadata: Optional[Metadata] | OpenAINotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | OpenAINotGiven = NOT_GIVEN,
+        n: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | OpenAINotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | OpenAINotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ChatCompletionReasoningEffort] | OpenAINotGiven = NOT_GIVEN,
+        thinking: ThinkingConfigParam | None | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | OpenAINotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | OpenAINotGiven = NOT_GIVEN,
+        store: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        user: str | OpenAINotGiven = NOT_GIVEN,
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | OpenAINotGiven = NOT_GIVEN,
     ):
         if model is not None:
             self.model = model
@@ -221,11 +318,6 @@ class OpenAICompatibleChatClient(BaseChatClient):
         else:
             self.response_format = {}
 
-        if stream_options:
-            _stream_options_params = {"stream_options": stream_options}
-        else:
-            _stream_options_params = {}
-
         self._acquire_rate_limit(self.endpoint, self.model, messages)
 
         if self.stream:
@@ -236,10 +328,33 @@ class OpenAICompatibleChatClient(BaseChatClient):
                 temperature=self.temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
-                **_stream_options_params,
-                **self.response_format,
-                **tools_params,
-                **kwargs,
+                audio=audio,
+                frequency_penalty=frequency_penalty,
+                logit_bias=logit_bias,
+                logprobs=logprobs,
+                max_completion_tokens=max_completion_tokens,
+                metadata=metadata,
+                modalities=modalities,
+                n=n,
+                parallel_tool_calls=parallel_tool_calls,
+                prediction=prediction,
+                presence_penalty=presence_penalty,
+                reasoning_effort=reasoning_effort,
+                seed=seed,
+                service_tier=service_tier,
+                stop=stop,
+                store=store,
+                top_logprobs=top_logprobs,
+                user=user,
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                stream_options=stream_options,
+                response_format=response_format
+                if response_format and self.model_setting.response_format_available
+                else OPENAI_NOT_GIVEN,
+                **tools_params,  # type: ignore
             )
 
             def generator():
@@ -341,9 +456,33 @@ class OpenAICompatibleChatClient(BaseChatClient):
                 temperature=self.temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
-                **self.response_format,
-                **tools_params,
-                **kwargs,
+                audio=audio,
+                frequency_penalty=frequency_penalty,
+                logit_bias=logit_bias,
+                logprobs=logprobs,
+                max_completion_tokens=max_completion_tokens,
+                metadata=metadata,
+                modalities=modalities,
+                n=n,
+                parallel_tool_calls=parallel_tool_calls,
+                prediction=prediction,
+                presence_penalty=presence_penalty,
+                reasoning_effort=reasoning_effort,
+                seed=seed,
+                service_tier=service_tier,
+                stop=stop,
+                store=store,
+                top_logprobs=top_logprobs,
+                user=user,
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                stream_options=stream_options,
+                response_format=response_format
+                if response_format and self.model_setting.response_format_available
+                else OPENAI_NOT_GIVEN,
+                **tools_params,  # type: ignore
             )
 
             result = {
@@ -436,11 +575,33 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         max_tokens: int | None = None,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
-        response_format: dict | None = None,
+        response_format: ResponseFormat | NotGiven = NOT_GIVEN,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
         skip_cutoff: bool = False,
-        **kwargs,
+        audio: Optional[ChatCompletionAudioParam] | OpenAINotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | OpenAINotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        metadata: Optional[Metadata] | OpenAINotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | OpenAINotGiven = NOT_GIVEN,
+        n: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | OpenAINotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | OpenAINotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ChatCompletionReasoningEffort] | OpenAINotGiven = NOT_GIVEN,
+        thinking: ThinkingConfigParam | None | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | OpenAINotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | OpenAINotGiven = NOT_GIVEN,
+        store: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        user: str | OpenAINotGiven = NOT_GIVEN,
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | OpenAINotGiven = NOT_GIVEN,
     ) -> ChatCompletionMessage:
         pass
 
@@ -455,11 +616,33 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         max_tokens: int | None = None,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
-        response_format: dict | None = None,
+        response_format: ResponseFormat | NotGiven = NOT_GIVEN,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
         skip_cutoff: bool = False,
-        **kwargs,
+        audio: Optional[ChatCompletionAudioParam] | OpenAINotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | OpenAINotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        metadata: Optional[Metadata] | OpenAINotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | OpenAINotGiven = NOT_GIVEN,
+        n: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | OpenAINotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | OpenAINotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ChatCompletionReasoningEffort] | OpenAINotGiven = NOT_GIVEN,
+        thinking: ThinkingConfigParam | None | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | OpenAINotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | OpenAINotGiven = NOT_GIVEN,
+        store: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        user: str | OpenAINotGiven = NOT_GIVEN,
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | OpenAINotGiven = NOT_GIVEN,
     ) -> AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
 
@@ -474,11 +657,33 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         max_tokens: int | None = None,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
-        response_format: dict | None = None,
+        response_format: ResponseFormat | NotGiven = NOT_GIVEN,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
         skip_cutoff: bool = False,
-        **kwargs,
+        audio: Optional[ChatCompletionAudioParam] | OpenAINotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | OpenAINotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        metadata: Optional[Metadata] | OpenAINotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | OpenAINotGiven = NOT_GIVEN,
+        n: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | OpenAINotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | OpenAINotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ChatCompletionReasoningEffort] | OpenAINotGiven = NOT_GIVEN,
+        thinking: ThinkingConfigParam | None | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | OpenAINotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | OpenAINotGiven = NOT_GIVEN,
+        store: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        user: str | OpenAINotGiven = NOT_GIVEN,
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | OpenAINotGiven = NOT_GIVEN,
     ) -> ChatCompletionMessage | AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
 
@@ -492,11 +697,33 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         max_tokens: int | None = None,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
-        response_format: dict | None = None,
+        response_format: ResponseFormat | NotGiven = NOT_GIVEN,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
         skip_cutoff: bool = False,
-        **kwargs,
+        audio: Optional[ChatCompletionAudioParam] | OpenAINotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | OpenAINotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        metadata: Optional[Metadata] | OpenAINotGiven = NOT_GIVEN,
+        modalities: Optional[List[ChatCompletionModality]] | OpenAINotGiven = NOT_GIVEN,
+        n: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | OpenAINotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | OpenAINotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | OpenAINotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ChatCompletionReasoningEffort] | OpenAINotGiven = NOT_GIVEN,
+        thinking: ThinkingConfigParam | None | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | OpenAINotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | OpenAINotGiven = NOT_GIVEN,
+        store: Optional[bool] | OpenAINotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | OpenAINotGiven = NOT_GIVEN,
+        user: str | OpenAINotGiven = NOT_GIVEN,
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | OpenAINotGiven = NOT_GIVEN,
     ):
         if model is not None:
             self.model = model
@@ -579,10 +806,33 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
                 temperature=self.temperature,
                 max_tokens=max_tokens,  # Azure 的 OpenAI 怎么 stream 模式不支持 max_completion_tokens
                 top_p=top_p,
-                **_stream_options_params,
-                **self.response_format,
-                **tools_params,
-                **kwargs,
+                audio=audio,
+                frequency_penalty=frequency_penalty,
+                logit_bias=logit_bias,
+                logprobs=logprobs,
+                max_completion_tokens=max_completion_tokens,
+                metadata=metadata,
+                modalities=modalities,
+                n=n,
+                parallel_tool_calls=parallel_tool_calls,
+                prediction=prediction,
+                presence_penalty=presence_penalty,
+                reasoning_effort=reasoning_effort,
+                seed=seed,
+                service_tier=service_tier,
+                stop=stop,
+                store=store,
+                top_logprobs=top_logprobs,
+                user=user,
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                stream_options=stream_options,
+                response_format=response_format
+                if response_format and self.model_setting.response_format_available
+                else OPENAI_NOT_GIVEN,
+                **tools_params,  # type: ignore
             )
 
             async def generator():
@@ -684,9 +934,33 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
                 temperature=self.temperature,
                 max_tokens=max_tokens,
                 top_p=top_p,
-                **self.response_format,
-                **tools_params,
-                **kwargs,
+                audio=audio,
+                frequency_penalty=frequency_penalty,
+                logit_bias=logit_bias,
+                logprobs=logprobs,
+                max_completion_tokens=max_completion_tokens,
+                metadata=metadata,
+                modalities=modalities,
+                n=n,
+                parallel_tool_calls=parallel_tool_calls,
+                prediction=prediction,
+                presence_penalty=presence_penalty,
+                reasoning_effort=reasoning_effort,
+                seed=seed,
+                service_tier=service_tier,
+                stop=stop,
+                store=store,
+                top_logprobs=top_logprobs,
+                user=user,
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                stream_options=stream_options,
+                response_format=response_format
+                if response_format and self.model_setting.response_format_available
+                else OPENAI_NOT_GIVEN,
+                **tools_params,  # type: ignore
             )
             result = {
                 "content": response.choices[0].message.content,

vectorvein/types/llm_parameters.py

@@ -5,12 +5,15 @@ from typing_extensions import TypedDict, NotRequired  # Required by pydantic und
 
 from pydantic import BaseModel, Field
 
-from anthropic.types import ToolParam as AnthropicToolParam
 from anthropic._types import NotGiven as AnthropicNotGiven
+from anthropic._types import NOT_GIVEN as ANTHROPIC_NOT_GIVEN
+from anthropic.types import ToolParam as AnthropicToolParam
+from anthropic.types import ThinkingConfigParam, ThinkingConfigEnabledParam
 from anthropic.types.message_create_params import ToolChoice as AnthropicToolChoice
 
 from openai._types import NotGiven as OpenAINotGiven
 from openai._types import NOT_GIVEN as OPENAI_NOT_GIVEN
+from openai.types.chat.completion_create_params import ResponseFormat
 from openai.types.chat.chat_completion_chunk import ChoiceDeltaToolCall
 from openai.types.chat.chat_completion_tool_param import ChatCompletionToolParam
 from openai.types.completion_usage import CompletionTokensDetails, PromptTokensDetails
@@ -150,4 +153,9 @@ __all__ = [
     "ToolChoice",
     "AnthropicToolParam",
     "AnthropicToolChoice",
+    "OPENAI_NOT_GIVEN",
+    "ANTHROPIC_NOT_GIVEN",
+    "ResponseFormat",
+    "ThinkingConfigParam",
+    "ThinkingConfigEnabledParam",
 ]

{vectorvein-0.2.7.dist-info → vectorvein-0.2.9.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.2.7
+Version: 0.2.9
 Summary: VectorVein python SDK
 Author-Email: Anderson <andersonby@163.com>
 License: MIT
@@ -8,7 +8,7 @@ Requires-Python: >=3.10
 Requires-Dist: openai>=1.37.1
 Requires-Dist: tiktoken>=0.7.0
 Requires-Dist: httpx>=0.27.0
-Requires-Dist: anthropic[bedrock,vertex]>=0.31.2
+Requires-Dist: anthropic[bedrock,vertex]>=0.47.1
 Requires-Dist: pydantic>=2.8.2
 Requires-Dist: Pillow>=10.4.0
 Requires-Dist: deepseek-tokenizer>=0.1.0

{vectorvein-0.2.7.dist-info → vectorvein-0.2.9.dist-info}/RECORD

@@ -1,15 +1,15 @@
-vectorvein-0.2.7.dist-info/METADATA,sha256=J67JgDpuGe0YZmLAxOKdkMYKaXD_6LjhSjAzmkzAfSM,4413
-vectorvein-0.2.7.dist-info/WHEEL,sha256=thaaA2w1JzcGC48WYufAs8nrYZjJm8LqNfnXFOFyCC4,90
-vectorvein-0.2.7.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
+vectorvein-0.2.9.dist-info/METADATA,sha256=0-oCeh7DA-fSPf3gKR0i79tcp_qLi5iOf1w4-mW9WaU,4413
+vectorvein-0.2.9.dist-info/WHEEL,sha256=thaaA2w1JzcGC48WYufAs8nrYZjJm8LqNfnXFOFyCC4,90
+vectorvein-0.2.9.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
 vectorvein/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 vectorvein/api/__init__.py,sha256=lfY-XA46fgD2iIZTU0VYP8i07AwA03Egj4Qua0vUKrQ,738
 vectorvein/api/client.py,sha256=xF-leKDQzVyyy9FnIRaz0k4eElYW1XbbzeRLcpnyk90,33047
 vectorvein/api/exceptions.py,sha256=uS_PAdx0ksC0r3dgfSGWdbLMZm4qdLeWSSqCv1g3_Gc,772
 vectorvein/api/models.py,sha256=xtPWMsB0yIJI7i-gY4B6MtvXv0ZIXnoeKspmeInH6fU,1449
-vectorvein/chat_clients/__init__.py,sha256=omQuG4PRRPNflSAgtdU--rwsWG6vMpwMEyIGZyFVHVQ,18596
-vectorvein/chat_clients/anthropic_client.py,sha256=8ns9oVFbGoRMWnizmskuDDW2C-xrbVVV_66s3uuhpNM,40079
+vectorvein/chat_clients/__init__.py,sha256=1ElJPca32YqSxD4dWTHkTEWhiJE_N6J-e5ZGilpT7ks,19178
+vectorvein/chat_clients/anthropic_client.py,sha256=Yx0H2kW-XWOZId9Dnzke5eWWS8FqMf6OfgczZF7yeuM,58192
 vectorvein/chat_clients/baichuan_client.py,sha256=CVMvpgjdrZGv0BWnTOBD-f2ufZ3wq3496wqukumsAr4,526
-vectorvein/chat_clients/base_client.py,sha256=QLvcGhjravPbvha6-spU-w6ugHU1LrsbdFUcs6NwMgE,18842
+vectorvein/chat_clients/base_client.py,sha256=RcAJwwJ83qC73t3W7GoPNvQXb-qPJ7d8U1J5nwzcYls,37044
 vectorvein/chat_clients/deepseek_client.py,sha256=3qWu01NlJAP2N-Ff62d5-CZXZitlizE1fzb20LNetig,526
 vectorvein/chat_clients/gemini_client.py,sha256=ufovIZrmAE3RLEe8h5WXombf7bARAZxnkj6ydNK2FQM,475
 vectorvein/chat_clients/groq_client.py,sha256=Uow4pgdmFi93ZQSoOol2-0PhhqkW-S0XuSldvppz5U4,498
@@ -18,7 +18,7 @@ vectorvein/chat_clients/minimax_client.py,sha256=YOILWcsHsN5tihLTMbKJIyJr9TJREMI
 vectorvein/chat_clients/mistral_client.py,sha256=1aKSylzBDaLYcFnaBIL4-sXSzWmXfBeON9Q0rq-ziWw,534
 vectorvein/chat_clients/moonshot_client.py,sha256=gbu-6nGxx8uM_U2WlI4Wus881rFRotzHtMSoYOcruGU,526
 vectorvein/chat_clients/openai_client.py,sha256=Nz6tV45pWcsOupxjnsRsGTicbQNJWIZyxuJoJ5DGMpg,527
-vectorvein/chat_clients/openai_compatible_client.py,sha256=H20SXdvXalf7_jaashff_aLCa8lO-n4-biBEY-6Nqfo,30572
+vectorvein/chat_clients/openai_compatible_client.py,sha256=IY3EKkkAespVeVV8sv7M1DN71BCdBRWm_dOtPLn2Rgo,48071
 vectorvein/chat_clients/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 vectorvein/chat_clients/qwen_client.py,sha256=-ryh-m9PgsO0fc4ulcCmPTy1155J8YUy15uPoJQOHA0,513
 vectorvein/chat_clients/stepfun_client.py,sha256=zsD2W5ahmR4DD9cqQTXmJr3txrGuvxbRWhFlRdwNijI,519
@@ -33,7 +33,7 @@ vectorvein/settings/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,
 vectorvein/types/defaults.py,sha256=VrkQoyHqC_eK3g1b6egpPYLLo0ltwMHqxDscCX4y-N0,27417
 vectorvein/types/enums.py,sha256=7KTJSVtQueImmbr1fSwv3rQVtc0RyMWXJmoE2tDOaso,1667
 vectorvein/types/exception.py,sha256=gnW4GnJ76jND6UGnodk9xmqkcbeS7Cz2rvncA2HpD5E,69
-vectorvein/types/llm_parameters.py,sha256=9KJ8z_7xVa0JsV0YPxRKz5n8Eb8PfvB9Y89e6ahjmjw,5989
+vectorvein/types/llm_parameters.py,sha256=uby71bteIHmZ3YtFBDbapOaQunBVqcI5qAfQmjWlniM,6339
 vectorvein/types/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 vectorvein/utilities/media_processing.py,sha256=7KtbLFzOYIn1e9QTN9G6C76NH8CBlV9kfAgiRKEIeXY,6263
 vectorvein/utilities/rate_limiter.py,sha256=dwolIUVw2wP83Odqpx0AAaE77de1GzxkYDGH4tM_u_4,10300
@@ -59,4 +59,4 @@ vectorvein/workflow/nodes/vector_db.py,sha256=t6I17q6iR3yQreiDHpRrksMdWDPIvgqJs0
 vectorvein/workflow/nodes/video_generation.py,sha256=qmdg-t_idpxq1veukd-jv_ChICMOoInKxprV9Z4Vi2w,4118
 vectorvein/workflow/nodes/web_crawlers.py,sha256=LsqomfXfqrXfHJDO1cl0Ox48f4St7X_SL12DSbAMSOw,5415
 vectorvein/workflow/utils/json_to_code.py,sha256=F7dhDy8kGc8ndOeihGLRLGFGlquoxVlb02ENtxnQ0C8,5914
-vectorvein-0.2.7.dist-info/RECORD,,
+vectorvein-0.2.9.dist-info/RECORD,,