vectorvein 0.1.58__tar.gz → 0.1.60__tar.gz

{vectorvein-0.1.58 → vectorvein-0.1.60}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.1.58
+Version: 0.1.60
 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[vertex]>=0.31.2
+Requires-Dist: anthropic[bedrock,vertex]>=0.31.2
 Requires-Dist: pydantic>=2.8.2
 Requires-Dist: Pillow>=10.4.0
 Requires-Dist: deepseek-tokenizer>=0.1.0

{vectorvein-0.1.58 → vectorvein-0.1.60}/pyproject.toml

@@ -6,7 +6,7 @@ dependencies = [
     "openai>=1.37.1",
     "tiktoken>=0.7.0",
     "httpx>=0.27.0",
-    "anthropic[vertex]>=0.31.2",
+    "anthropic[vertex,bedrock]>=0.31.2",
     "pydantic>=2.8.2",
     "Pillow>=10.4.0",
     "deepseek-tokenizer>=0.1.0",
@@ -17,7 +17,7 @@ description = "VectorVein python SDK"
 name = "vectorvein"
 readme = "README.md"
 requires-python = ">=3.10"
-version = "0.1.58"
+version = "0.1.60"
 
 [project.license]
 text = "MIT"

{vectorvein-0.1.58 → vectorvein-0.1.60}/src/vectorvein/chat_clients/anthropic_client.py

@@ -8,7 +8,14 @@ from typing import overload, Generator, AsyncGenerator, Any, Literal, Iterable
 import httpx
 from openai._types import NotGiven as OpenAINotGiven
 from openai._types import NOT_GIVEN as OPENAI_NOT_GIVEN
-from anthropic import Anthropic, AnthropicVertex, AsyncAnthropic, AsyncAnthropicVertex
+from anthropic import (
+    Anthropic,
+    AnthropicVertex,
+    AsyncAnthropic,
+    AsyncAnthropicVertex,
+    AnthropicBedrock,
+    AsyncAnthropicBedrock,
+)
 from anthropic._types import NOT_GIVEN
 from anthropic.types import (
     TextBlock,
@@ -219,6 +226,15 @@ class AnthropicChatClient(BaseChatClient):
                 access_token=self.creds.token,
                 http_client=self.http_client,
             )
+        elif self.endpoint.is_bedrock:
+            if self.endpoint.credentials is None:
+                raise ValueError("Anthropic Bedrock endpoint requires credentials")
+            return AnthropicBedrock(
+                aws_access_key=self.endpoint.credentials.get("access_key"),
+                aws_secret_key=self.endpoint.credentials.get("secret_key"),
+                aws_region=self.endpoint.region,
+                http_client=self.http_client,
+            )
         elif self.endpoint.api_schema_type == "default":
             return Anthropic(
                 api_key=self.endpoint.api_key,
@@ -250,6 +266,7 @@ class AnthropicChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -267,6 +284,7 @@ class AnthropicChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> Generator[ChatCompletionDeltaMessage, None, None]:
         pass
@@ -284,6 +302,7 @@ class AnthropicChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | Generator[ChatCompletionDeltaMessage, Any, None]:
         pass
@@ -300,6 +319,7 @@ class AnthropicChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ):
         if model is not None:
@@ -349,6 +369,7 @@ class AnthropicChatClient(BaseChatClient):
                         response_format=response_format,
                         stream_options=stream_options,
                         top_p=top_p,
+                        skip_cutoff=skip_cutoff,
                         **kwargs,
                     )
                     for chunk in response:
@@ -374,8 +395,8 @@ class AnthropicChatClient(BaseChatClient):
                     tools=_tools,
                     tool_choice=_tool_choice,
                     response_format=response_format,
-                    stream_options=stream_options,
                     top_p=top_p,
+                    skip_cutoff=skip_cutoff,
                     **kwargs,
                 )
 
@@ -399,7 +420,7 @@ class AnthropicChatClient(BaseChatClient):
         else:
             system_prompt = ""
 
-        if self.context_length_control == ContextLengthControlType.Latest:
+        if not skip_cutoff and self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
                 max_count=self.model_setting.context_length,
@@ -595,6 +616,15 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
                 access_token=self.creds.token,
                 http_client=self.http_client,
             )
+        elif self.endpoint.is_bedrock:
+            if self.endpoint.credentials is None:
+                raise ValueError("Anthropic Bedrock endpoint requires credentials")
+            return AsyncAnthropicBedrock(
+                aws_access_key=self.endpoint.credentials.get("aws_access_key"),
+                aws_secret_key=self.endpoint.credentials.get("aws_secret_key"),
+                aws_region=self.endpoint.region,
+                http_client=self.http_client,
+            )
         elif self.endpoint.api_schema_type == "default":
             return AsyncAnthropic(
                 api_key=self.endpoint.api_key,
@@ -626,6 +656,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -643,6 +674,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
@@ -660,6 +692,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
@@ -676,6 +709,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ):
         if model is not None:
@@ -725,6 +759,8 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
                         tool_choice=_tool_choice,
                         response_format=response_format,
                         stream_options=stream_options,
+                        top_p=top_p,
+                        skip_cutoff=skip_cutoff,
                         **kwargs,
                     )
                     async for chunk in response:
@@ -751,6 +787,8 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
                     tools=_tools,
                     tool_choice=_tool_choice,
                     response_format=response_format,
+                    top_p=top_p,
+                    skip_cutoff=skip_cutoff,
                     **kwargs,
                 )
 
@@ -774,7 +812,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
         else:
             system_prompt = ""
 
-        if self.context_length_control == ContextLengthControlType.Latest:
+        if not skip_cutoff and self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
                 max_count=self.model_setting.context_length,

{vectorvein-0.1.58 → vectorvein-0.1.60}/src/vectorvein/chat_clients/base_client.py

@@ -6,7 +6,14 @@ from typing import Generator, AsyncGenerator, Any, overload, Literal, Iterable
 
 import httpx
 from openai import OpenAI, AsyncOpenAI, AzureOpenAI, AsyncAzureOpenAI
-from anthropic import Anthropic, AnthropicVertex, AsyncAnthropic, AsyncAnthropicVertex
+from anthropic import (
+    Anthropic,
+    AnthropicVertex,
+    AsyncAnthropic,
+    AsyncAnthropicVertex,
+    AnthropicBedrock,
+    AsyncAnthropicBedrock,
+)
 
 from ..settings import settings
 from ..types import defaults as defs
@@ -57,7 +64,9 @@ class BaseChatClient(ABC):
 
     @cached_property
     @abstractmethod
-    def raw_client(self) -> OpenAI | AzureOpenAI | Anthropic | AnthropicVertex | httpx.Client | None:
+    def raw_client(
+        self,
+    ) -> OpenAI | AzureOpenAI | Anthropic | AnthropicVertex | AnthropicBedrock | httpx.Client | None:
         pass
 
     @overload
@@ -74,6 +83,7 @@ class BaseChatClient(ABC):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -92,6 +102,7 @@ class BaseChatClient(ABC):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> Generator[ChatCompletionDeltaMessage, Any, None]:
         pass
@@ -110,6 +121,7 @@ class BaseChatClient(ABC):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | Generator[ChatCompletionDeltaMessage, Any, None]:
         pass
@@ -127,6 +139,7 @@ class BaseChatClient(ABC):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | Generator[ChatCompletionDeltaMessage, Any, None]:
         pass
@@ -195,7 +208,15 @@ class BaseAsyncChatClient(ABC):
     @abstractmethod
     def raw_client(
         self,
-    ) -> AsyncOpenAI | AsyncAzureOpenAI | AsyncAnthropic | AsyncAnthropicVertex | httpx.AsyncClient | None:
+    ) -> (
+        AsyncOpenAI
+        | AsyncAzureOpenAI
+        | AsyncAnthropic
+        | AsyncAnthropicVertex
+        | AsyncAnthropicBedrock
+        | httpx.AsyncClient
+        | None
+    ):
         pass
 
     @overload
@@ -212,6 +233,7 @@ class BaseAsyncChatClient(ABC):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -230,6 +252,7 @@ class BaseAsyncChatClient(ABC):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> AsyncGenerator[ChatCompletionDeltaMessage, None]:
         pass
@@ -248,6 +271,7 @@ class BaseAsyncChatClient(ABC):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | AsyncGenerator[ChatCompletionDeltaMessage, None]:
         pass
@@ -265,6 +289,7 @@ class BaseAsyncChatClient(ABC):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | AsyncGenerator[ChatCompletionDeltaMessage, None]:
         pass

{vectorvein-0.1.58 → vectorvein-0.1.60}/src/vectorvein/chat_clients/gemini_client.py

@@ -66,6 +66,7 @@ class GeminiChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -83,6 +84,7 @@ class GeminiChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> Generator[ChatCompletionDeltaMessage, None, None]:
         pass
@@ -100,6 +102,7 @@ class GeminiChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | Generator[ChatCompletionDeltaMessage, Any, None]:
         pass
@@ -116,6 +119,7 @@ class GeminiChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ):
         if model is not None:
@@ -133,7 +137,7 @@ class GeminiChatClient(BaseChatClient):
         else:
             system_prompt = ""
 
-        if self.context_length_control == ContextLengthControlType.Latest:
+        if not skip_cutoff and self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
                 max_count=self.model_setting.context_length,
@@ -313,6 +317,7 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -330,6 +335,7 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
@@ -347,6 +353,7 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
@@ -363,6 +370,7 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ):
         if model is not None:
@@ -380,7 +388,7 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
         else:
             system_prompt = ""
 
-        if self.context_length_control == ContextLengthControlType.Latest:
+        if not skip_cutoff and self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
                 max_count=self.model_setting.context_length,

{vectorvein-0.1.58 → vectorvein-0.1.60}/src/vectorvein/chat_clients/minimax_client.py

@@ -92,6 +92,7 @@ class MiniMaxChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -109,6 +110,7 @@ class MiniMaxChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> Generator[ChatCompletionDeltaMessage, None, None]:
         pass
@@ -126,6 +128,7 @@ class MiniMaxChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | Generator[ChatCompletionDeltaMessage, Any, None]:
         pass
@@ -142,6 +145,7 @@ class MiniMaxChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ):
         if model is not None:
@@ -166,7 +170,7 @@ class MiniMaxChatClient(BaseChatClient):
                 self.endpoint_id = endpoint_choice
             self.endpoint = settings.get_endpoint(self.endpoint_id)
 
-        if self.context_length_control == ContextLengthControlType.Latest:
+        if not skip_cutoff and self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
                 max_count=self.model_setting.context_length,
@@ -337,6 +341,7 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -354,6 +359,7 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
@@ -371,6 +377,7 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
@@ -387,6 +394,7 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None = None,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ):
         if model is not None:
@@ -411,7 +419,7 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
                 self.endpoint_id = endpoint_choice
             self.endpoint = settings.get_endpoint(self.endpoint_id)
 
-        if self.context_length_control == ContextLengthControlType.Latest:
+        if not skip_cutoff and self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
                 max_count=self.model_setting.context_length,

{vectorvein-0.1.58 → vectorvein-0.1.60}/src/vectorvein/chat_clients/openai_compatible_client.py

@@ -99,6 +99,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -116,6 +117,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> Generator[ChatCompletionDeltaMessage, None, None]:
         pass
@@ -133,6 +135,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | Generator[ChatCompletionDeltaMessage, Any, None]:
         pass
@@ -149,6 +152,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ):
         if model is not None:
@@ -165,7 +169,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         if self.model_id is None:
             self.model_id = self.model_setting.id
 
-        if self.context_length_control == ContextLengthControlType.Latest:
+        if not skip_cutoff and self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
                 max_count=self.model_setting.context_length,
@@ -361,6 +365,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage:
         pass
@@ -378,6 +383,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
@@ -395,6 +401,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ) -> ChatCompletionMessage | AsyncGenerator[ChatCompletionDeltaMessage, Any]:
         pass
@@ -411,6 +418,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         response_format: dict | None = None,
         stream_options: ChatCompletionStreamOptionsParam | None | OpenAINotGiven = NOT_GIVEN,
         top_p: float | NotGiven | None = NOT_GIVEN,
+        skip_cutoff: bool = False,
         **kwargs,
     ):
         if model is not None:
@@ -427,7 +435,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         if self.model_id is None:
             self.model_id = self.model_setting.id
 
-        if self.context_length_control == ContextLengthControlType.Latest:
+        if not skip_cutoff and self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
                 max_count=self.model_setting.context_length,

{vectorvein-0.1.58 → vectorvein-0.1.60}/src/vectorvein/types/defaults.py

@@ -231,6 +231,20 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": False,
         "response_format_available": True,
     },
+    "qwen2.5-coder-32b-instruct": {
+        "id": "qwen2.5-coder-32b-instruct",
+        "context_length": 30000,
+        "max_output_tokens": 4096,
+        "function_call_available": False,
+        "response_format_available": False,
+    },
+    "qwq-32b-preview": {
+        "id": "qwq-32b-preview",
+        "context_length": 30000,
+        "max_output_tokens": 4096,
+        "function_call_available": False,
+        "response_format_available": False,
+    },
     "qwen2.5-72b-instruct": {
         "id": "qwen2.5-72b-instruct",
         "context_length": 131072,
@@ -238,6 +252,14 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": False,
         "response_format_available": True,
     },
+    "qwen2-vl-72b-instruct": {
+        "id": "qwen2-vl-72b-instruct",
+        "context_length": 131072,
+        "max_output_tokens": 8192,
+        "function_call_available": False,
+        "response_format_available": False,
+        "native_multimodal": True,
+    },
     "qwen-max": {
         "id": "qwen-max",
         "context_length": 8096,

{vectorvein-0.1.58 → vectorvein-0.1.60}/src/vectorvein/types/llm_parameters.py

@@ -37,6 +37,7 @@ class EndpointSetting(BaseModel):
     credentials: Optional[dict] = Field(None, description="Additional credentials if needed.")
     is_azure: bool = Field(False, description="Indicates if the endpoint is for Azure.")
     is_vertex: bool = Field(False, description="Indicates if the endpoint is for Vertex.")
+    is_bedrock: bool = Field(False, description="Indicates if the endpoint is for Bedrock.")
     rpm: int = Field(description="Requests per minute.", default=defs.ENDPOINT_RPM)
     tpm: int = Field(description="Tokens per minute.", default=defs.ENDPOINT_TPM)
     concurrent_requests: int = Field(