vectorvein 0.1.11__py3-none-any.whl → 0.1.12__py3-none-any.whl

vectorvein/chat_clients/__init__.py

@@ -1,5 +1,7 @@
 # @Author: Bi Ying
 # @Date:   2024-07-26 14:48:55
+import httpx
+
 from .base_client import BaseChatClient, BaseAsyncChatClient
 
 from .yi_client import YiChatClient, AsyncYiChatClient
@@ -58,6 +60,9 @@ def create_chat_client(
     stream: bool = False,
     temperature: float = 0.7,
     context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
+    random_endpoint: bool = True,
+    endpoint_id: str = "",
+    http_client: httpx.Client | None = None,
     **kwargs,
 ) -> BaseChatClient:
     if backend.lower() not in BackendMap["sync"]:
@@ -73,6 +78,9 @@ def create_chat_client(
         stream=stream,
         temperature=temperature,
         context_length_control=context_length_control,
+        random_endpoint=random_endpoint,
+        endpoint_id=endpoint_id,
+        http_client=http_client,
         **kwargs,
     )
 
@@ -83,6 +91,9 @@ def create_async_chat_client(
     stream: bool = False,
     temperature: float = 0.7,
     context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
+    random_endpoint: bool = True,
+    endpoint_id: str = "",
+    http_client: httpx.AsyncClient | None = None,
     **kwargs,
 ) -> BaseAsyncChatClient:
     if backend.lower() not in BackendMap["async"]:
@@ -98,6 +109,9 @@ def create_async_chat_client(
         stream=stream,
         temperature=temperature,
         context_length_control=context_length_control,
+        random_endpoint=random_endpoint,
+        endpoint_id=endpoint_id,
+        http_client=http_client,
         **kwargs,
     )
 

vectorvein/chat_clients/anthropic_client.py

@@ -3,6 +3,7 @@
 import json
 import random
 
+import httpx
 from openai._types import NotGiven as OpenAINotGiven
 from anthropic import Anthropic, AnthropicVertex, AsyncAnthropic, AsyncAnthropicVertex
 from anthropic._types import NotGiven, NOT_GIVEN
@@ -97,6 +98,7 @@ class AnthropicChatClient(BaseChatClient):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.Client | None = None,
         **kwargs,
     ):
         super().__init__(
@@ -106,6 +108,7 @@ class AnthropicChatClient(BaseChatClient):
             context_length_control,
             random_endpoint,
             endpoint_id,
+            http_client,
             **kwargs,
         )
 
@@ -183,11 +186,13 @@ class AnthropicChatClient(BaseChatClient):
                 base_url=base_url,
                 project_id=self.endpoint.credentials.get("quota_project_id"),
                 access_token=self.creds.token,
+                http_client=self.http_client,
             )
         else:
             self._client = Anthropic(
                 api_key=self.endpoint.api_key,
                 base_url=self.endpoint.api_base,
+                http_client=self.http_client,
             )
 
         tools_params = refactor_tool_use_params(tools) if tools else tools
@@ -299,6 +304,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.AsyncClient | None = None,
         **kwargs,
     ):
         super().__init__(
@@ -308,6 +314,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
             context_length_control,
             random_endpoint,
             endpoint_id,
+            http_client,
             **kwargs,
         )
 
@@ -385,11 +392,13 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
                 base_url=base_url,
                 project_id=self.endpoint.credentials.get("quota_project_id"),
                 access_token=self.creds.token,
+                http_client=self.http_client,
             )
         else:
             self._client = AsyncAnthropic(
                 api_key=self.endpoint.api_key,
                 base_url=self.endpoint.api_base,
+                http_client=self.http_client,
             )
 
         tools_params = refactor_tool_use_params(tools) if tools else tools

vectorvein/chat_clients/base_client.py

@@ -3,6 +3,7 @@
 from abc import ABC, abstractmethod
 from typing import Generator, AsyncGenerator, Any
 
+import httpx
 from openai._types import NotGiven, NOT_GIVEN
 
 from ..settings import settings
@@ -23,6 +24,7 @@ class BaseChatClient(ABC):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.Client | None = None,
         **kwargs,
     ):
         self.model = model or self.DEFAULT_MODEL
@@ -31,6 +33,7 @@ class BaseChatClient(ABC):
         self.context_length_control = context_length_control
         self.random_endpoint = random_endpoint
         self.endpoint_id = endpoint_id
+        self.http_client = http_client
 
         self.backend_settings = settings.get_backend(self.BACKEND_NAME)
 
@@ -84,6 +87,7 @@ class BaseAsyncChatClient(ABC):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.AsyncClient | None = None,
         **kwargs,
     ):
         self.model = model or self.DEFAULT_MODEL
@@ -92,6 +96,7 @@ class BaseAsyncChatClient(ABC):
         self.context_length_control = context_length_control
         self.random_endpoint = random_endpoint
         self.endpoint_id = endpoint_id
+        self.http_client = http_client
 
         self.backend_settings = settings.get_backend(self.BACKEND_NAME)
 

vectorvein/chat_clients/gemini_client.py

@@ -25,6 +25,7 @@ class GeminiChatClient(BaseChatClient):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.Client | None = None,
         **kwargs,
     ):
         super().__init__(
@@ -34,6 +35,7 @@ class GeminiChatClient(BaseChatClient):
             context_length_control,
             random_endpoint,
             endpoint_id,
+            http_client,
             **kwargs,
         )
 
@@ -107,7 +109,11 @@ class GeminiChatClient(BaseChatClient):
 
             def generator():
                 result = {"content": ""}
-                with httpx.stream("POST", url, headers=headers, params=params, json=request_body) as response:
+                if self.http_client:
+                    client = self.http_client
+                else:
+                    client = httpx.Client()
+                with client.stream("POST", url, headers=headers, params=params, json=request_body) as response:
                     for chunk in response.iter_lines():
                         message = {"content": ""}
                         if not chunk.startswith("data:"):
@@ -142,7 +148,11 @@ class GeminiChatClient(BaseChatClient):
             return generator()
         else:
             url = f"{self.endpoint.api_base}/models/{self.model_setting.id}:generateContent"
-            response = httpx.post(url, json=request_body, headers=headers, params=params, timeout=None).json()
+            if self.http_client:
+                client = self.http_client
+            else:
+                client = httpx.Client()
+            response = client.post(url, json=request_body, headers=headers, params=params, timeout=None).json()
             result = {
                 "content": "",
                 "usage": {
@@ -185,6 +195,7 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.AsyncClient | None = None,
         **kwargs,
     ):
         super().__init__(
@@ -194,6 +205,7 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
             context_length_control,
             random_endpoint,
             endpoint_id,
+            http_client,
             **kwargs,
         )
 
@@ -267,7 +279,10 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
 
             async def generator():
                 result = {"content": ""}
-                client = httpx.AsyncClient()
+                if self.http_client:
+                    client = self.http_client
+                else:
+                    client = httpx.AsyncClient()
                 async with client.stream("POST", url, headers=headers, params=params, json=request_body) as response:
                     async for chunk in response.aiter_lines():
                         message = {"content": ""}
@@ -303,8 +318,12 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
             return generator()
         else:
             url = f"{self.endpoint.api_base}/models/{self.model_setting.id}:generateContent"
-            async with httpx.AsyncClient(headers=headers, params=params, timeout=None) as client:
-                response = await client.post(url, json=request_body)
+            if self.http_client:
+                client = self.http_client
+            else:
+                client = httpx.AsyncClient()
+            async with client:
+                response = await client.post(url, json=request_body, headers=headers, params=params, timeout=None)
                 response = response.json()
                 result = {
                     "content": "",

vectorvein/chat_clients/minimax_client.py

@@ -48,6 +48,7 @@ class MiniMaxChatClient(BaseChatClient):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.Client | None = None,
         **kwargs,
     ):
         super().__init__(
@@ -57,9 +58,13 @@ class MiniMaxChatClient(BaseChatClient):
             context_length_control,
             random_endpoint,
             endpoint_id,
+            http_client,
             **kwargs,
         )
-        self.http_client = httpx.Client()
+        if http_client:
+            self.http_client = http_client
+        else:
+            self.http_client = httpx.Client()
 
     def create_completion(
         self,
@@ -208,6 +213,7 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.AsyncClient | None = None,
         **kwargs,
     ):
         super().__init__(
@@ -217,9 +223,13 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
             context_length_control,
             random_endpoint,
             endpoint_id,
+            http_client,
             **kwargs,
         )
-        self.http_client = httpx.AsyncClient()
+        if http_client:
+            self.http_client = http_client
+        else:
+            self.http_client = httpx.AsyncClient()
 
     async def create_completion(
         self,

vectorvein/chat_clients/openai_compatible_client.py

@@ -3,6 +3,7 @@
 import json
 import random
 
+import httpx
 from openai._types import NotGiven, NOT_GIVEN
 from openai._streaming import Stream, AsyncStream
 from openai.types.chat import ChatCompletion, ChatCompletionChunk
@@ -33,6 +34,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.Client | None = None,
         **kwargs,
     ):
         super().__init__(
@@ -42,6 +44,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
             context_length_control,
             random_endpoint,
             endpoint_id,
+            http_client,
             **kwargs,
         )
 
@@ -74,12 +77,14 @@ class OpenAICompatibleChatClient(BaseChatClient):
             self._client = AzureOpenAI(
                 azure_endpoint=self.endpoint.api_base,
                 api_key=self.endpoint.api_key,
-                api_version="2024-05-01-preview",
+                api_version="2024-08-01-preview",
+                http_client=self.http_client,
             )
         else:
             self._client = OpenAI(
                 api_key=self.endpoint.api_key,
                 base_url=self.endpoint.api_base,
+                http_client=self.http_client,
             )
 
         if self.context_length_control == ContextLengthControlType.Latest:
@@ -188,6 +193,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         context_length_control: ContextLengthControlType = defs.CONTEXT_LENGTH_CONTROL,
         random_endpoint: bool = True,
         endpoint_id: str = "",
+        http_client: httpx.AsyncClient | None = None,
         **kwargs,
     ):
         super().__init__(
@@ -197,6 +203,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
             context_length_control,
             random_endpoint,
             endpoint_id,
+            http_client,
             **kwargs,
         )
 
@@ -229,12 +236,14 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
             self._client = AsyncAzureOpenAI(
                 azure_endpoint=self.endpoint.api_base,
                 api_key=self.endpoint.api_key,
-                api_version="2024-05-01-preview",
+                api_version="2024-08-01-preview",
+                http_client=self.http_client,
             )
         else:
             self._client = AsyncOpenAI(
                 api_key=self.endpoint.api_key,
                 base_url=self.endpoint.api_base,
+                http_client=self.http_client,
             )
 
         if self.context_length_control == ContextLengthControlType.Latest:

vectorvein/types/llm_parameters.py

@@ -69,6 +69,7 @@ class ChatCompletionMessage(BaseModel):
 
     usage: Optional[Usage] = None
 
+
 class ChatCompletionDeltaMessage(BaseModel):
     content: Optional[str] = None
 

{vectorvein-0.1.11.dist-info → vectorvein-0.1.12.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.1.11
+Version: 0.1.12
 Summary: Default template for PDM package
 Author-Email: Anderson <andersonby@163.com>
 License: MIT

{vectorvein-0.1.11.dist-info → vectorvein-0.1.12.dist-info}/RECORD

@@ -1,18 +1,18 @@
-vectorvein-0.1.11.dist-info/METADATA,sha256=rw0fFY-7V-teJUIt4t4KlvLIZHp4hjgaQBRpQ816eP0,502
-vectorvein-0.1.11.dist-info/WHEEL,sha256=rSwsxJWe3vzyR5HCwjWXQruDgschpei4h_giTm0dJVE,90
+vectorvein-0.1.12.dist-info/METADATA,sha256=OxvgLVzQ5ilVePTHSj5ZhU9aKdyb7nXq7A7UNyDFuzo,502
+vectorvein-0.1.12.dist-info/WHEEL,sha256=rSwsxJWe3vzyR5HCwjWXQruDgschpei4h_giTm0dJVE,90
 vectorvein/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-vectorvein/chat_clients/__init__.py,sha256=5j7W--jr-l2cqDJp38uXYvkydDK0rnzm7MYGSACHKmU,3976
-vectorvein/chat_clients/anthropic_client.py,sha256=MHpLywbIIgFsAkj8EiKEoxUgxcbx3njFxLTN2GhEjys,20088
-vectorvein/chat_clients/base_client.py,sha256=wMXpQ1L1KDb2Hg6va3H3GmcVeQB6r6sh7F4IS0DBQWI,4275
+vectorvein/chat_clients/__init__.py,sha256=rur7elYPL-HIQM46U5Kb8aO4tMXfYOOKLdIHFlKIw0M,4421
+vectorvein/chat_clients/anthropic_client.py,sha256=sECXynbReJs1xixFEHSlfEQo9Q1zGf5uCJ-V8emdoAM,20438
+vectorvein/chat_clients/base_client.py,sha256=CscLWDNd8pE0EKthvRRtxyjMhGGt7pwP4FOts9KNOLU,4469
 vectorvein/chat_clients/deepseek_client.py,sha256=3qWu01NlJAP2N-Ff62d5-CZXZitlizE1fzb20LNetig,526
-vectorvein/chat_clients/gemini_client.py,sha256=jUu-wwpaChP8_lxLdWnYkgVr_aSegCHbmBgnpFztqAw,13743
+vectorvein/chat_clients/gemini_client.py,sha256=W-9Vu-GTE9wxStPznyNR0rBEgDG3LYBu2uQXd4sh1YQ,14425
 vectorvein/chat_clients/groq_client.py,sha256=Uow4pgdmFi93ZQSoOol2-0PhhqkW-S0XuSldvppz5U4,498
 vectorvein/chat_clients/local_client.py,sha256=55nOsxzqUf79q3Y14MKROA71zxhsT7p7FsDZ89rts2M,422
-vectorvein/chat_clients/minimax_client.py,sha256=ZO-7WNsvsct3cpuo0Qc6WL3A0kkijn0J8o0ahlxmUyE,13478
+vectorvein/chat_clients/minimax_client.py,sha256=iNq87zWx79g8tGo784c67xUt-YQ4TyMoFWUIjDVEMGw,13801
 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=rPxIPj4STm1sy7T6bqyZpO-FPwMyjyQmThyJqd8_jYs,13822
+vectorvein/chat_clients/openai_compatible_client.py,sha256=00Kxwmnooh-yfENkNLgl1Ped2s912njwVitqes8SZpw,14181
 vectorvein/chat_clients/qwen_client.py,sha256=-ryh-m9PgsO0fc4ulcCmPTy1155J8YUy15uPoJQOHA0,513
 vectorvein/chat_clients/utils.py,sha256=mnAew2Ie3nQHdEyDLKuJvXkQ5QdcSAJ6SpYk5JPbR1Q,20888
 vectorvein/chat_clients/yi_client.py,sha256=RNf4CRuPJfixrwLZ3-DEc3t25QDe1mvZeb9sku2f8Bc,484
@@ -20,7 +20,7 @@ vectorvein/chat_clients/zhipuai_client.py,sha256=Ys5DSeLCuedaDXr3PfG1EW2zKXopt-a
 vectorvein/settings/__init__.py,sha256=4mpccT7eZC3yI1vVnVViW4wHBnDEH9D2R5EsIP34VgU,3218
 vectorvein/types/defaults.py,sha256=lCzGOLybX8FzHX-Cv32BaQFZ8sHvPhGIIwDD-VksP20,13460
 vectorvein/types/enums.py,sha256=vzOenCnRlFXBwPh-lfFhjGfM-6yfDj7wZColHODqocI,1550
-vectorvein/types/llm_parameters.py,sha256=nBjStC2zndTY__yhD2WFXB09taxEhDLE3OHA6MICfgE,3494
+vectorvein/types/llm_parameters.py,sha256=PWN18dDGrCTP4Bz7pX0XxO-wDUA7qTngppzEELrROmc,3496
 vectorvein/utilities/media_processing.py,sha256=BujciRmw1GMmc3ELRvafL8STcy6r5b2rVnh27-uA7so,2256
 vectorvein/utilities/retry.py,sha256=9ePuJdeUUGx-qMWfaFxmlOvG_lQPwCQ4UB1z3Edlo34,993
-vectorvein-0.1.11.dist-info/RECORD,,
+vectorvein-0.1.12.dist-info/RECORD,,