vectorvein 0.2.67__py3-none-any.whl → 0.2.69__py3-none-any.whl

vectorvein/chat_clients/anthropic_client.py

@@ -766,7 +766,7 @@ class AnthropicChatClient(BaseChatClient):
                 "usage": {
                     "prompt_tokens": response.usage.input_tokens + response.usage.cache_read_input_tokens
                     if response.usage.cache_read_input_tokens
-                    else 0,
+                    else response.usage.input_tokens,
                     "completion_tokens": response.usage.output_tokens,
                     "total_tokens": response.usage.input_tokens + response.usage.output_tokens,
                     "prompt_tokens_details": {
@@ -1377,7 +1377,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
                 "usage": {
                     "prompt_tokens": response.usage.input_tokens + response.usage.cache_read_input_tokens
                     if response.usage.cache_read_input_tokens
-                    else 0,
+                    else response.usage.input_tokens,
                     "completion_tokens": response.usage.output_tokens,
                     "total_tokens": response.usage.input_tokens + response.usage.output_tokens,
                     "prompt_tokens_details": {

vectorvein/chat_clients/openai_compatible_client.py

@@ -117,7 +117,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         model: str | None = None,
         stream: Literal[False] = False,
         temperature: float | None | NotGiven = NOT_GIVEN,
-        max_tokens: int | None = None,
+        max_tokens: int | None | NotGiven = NOT_GIVEN,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
         response_format: ResponseFormat | NotGiven = NOT_GIVEN,
@@ -158,7 +158,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         model: str | None = None,
         stream: Literal[True],
         temperature: float | None | NotGiven = NOT_GIVEN,
-        max_tokens: int | None = None,
+        max_tokens: int | None | NotGiven = NOT_GIVEN,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
         response_format: ResponseFormat | NotGiven = NOT_GIVEN,
@@ -199,7 +199,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         model: str | None = None,
         stream: bool,
         temperature: float | None | NotGiven = NOT_GIVEN,
-        max_tokens: int | None = None,
+        max_tokens: int | None | NotGiven = NOT_GIVEN,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
         response_format: ResponseFormat | NotGiven = NOT_GIVEN,
@@ -239,7 +239,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         model: str | None = None,
         stream: Literal[False] | Literal[True] = False,
         temperature: float | None | NotGiven = NOT_GIVEN,
-        max_tokens: int | None = None,
+        max_tokens: int | None | NotGiven = NOT_GIVEN,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
         response_format: ResponseFormat | NotGiven = NOT_GIVEN,
@@ -280,6 +280,8 @@ class OpenAICompatibleChatClient(BaseChatClient):
             self.temperature = temperature
         if isinstance(top_p, AnthropicNotGiven):
             top_p = NOT_GIVEN
+        if isinstance(max_tokens, AnthropicNotGiven):
+            max_tokens = NOT_GIVEN
 
         raw_client = self.raw_client  # 调用完 self.raw_client 后，self.model_id 会被赋值
         self.model_setting = self.backend_settings.models[self.model]
@@ -322,7 +324,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         else:
             tools_params = {}
 
-        if max_tokens is None:
+        if not max_tokens and not max_completion_tokens:
             max_output_tokens = self.model_setting.max_output_tokens
             token_counts = get_message_token_counts(messages=messages, tools=tools, model=self.model)
             if max_output_tokens is not None:
@@ -331,6 +333,10 @@ class OpenAICompatibleChatClient(BaseChatClient):
             else:
                 max_tokens = self.model_setting.context_length - token_counts - 64
 
+        if "o3-mini" in self.model_id:
+            max_completion_tokens = max_tokens
+            max_tokens = NOT_GIVEN
+
         self._acquire_rate_limit(self.endpoint, self.model, messages)
 
         if self.stream:
@@ -587,7 +593,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         model: str | None = None,
         stream: Literal[False] = False,
         temperature: float | None | NotGiven = NOT_GIVEN,
-        max_tokens: int | None = None,
+        max_tokens: int | None | NotGiven = NOT_GIVEN,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
         response_format: ResponseFormat | NotGiven = NOT_GIVEN,
@@ -628,7 +634,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         model: str | None = None,
         stream: Literal[True],
         temperature: float | None | NotGiven = NOT_GIVEN,
-        max_tokens: int | None = None,
+        max_tokens: int | None | NotGiven = NOT_GIVEN,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
         response_format: ResponseFormat | NotGiven = NOT_GIVEN,
@@ -669,7 +675,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         model: str | None = None,
         stream: bool,
         temperature: float | None | NotGiven = NOT_GIVEN,
-        max_tokens: int | None = None,
+        max_tokens: int | None | NotGiven = NOT_GIVEN,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
         response_format: ResponseFormat | NotGiven = NOT_GIVEN,
@@ -709,7 +715,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         model: str | None = None,
         stream: Literal[False] | Literal[True] = False,
         temperature: float | None | NotGiven = NOT_GIVEN,
-        max_tokens: int | None = None,
+        max_tokens: int | None | NotGiven = NOT_GIVEN,
         tools: Iterable[ToolParam] | NotGiven = NOT_GIVEN,
         tool_choice: ToolChoice | NotGiven = NOT_GIVEN,
         response_format: ResponseFormat | NotGiven = NOT_GIVEN,
@@ -750,6 +756,8 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
             self.temperature = temperature
         if isinstance(top_p, AnthropicNotGiven):
             top_p = NOT_GIVEN
+        if isinstance(max_tokens, AnthropicNotGiven):
+            max_tokens = NOT_GIVEN
 
         raw_client = self.raw_client  # 调用完 self.raw_client 后，self.model_id 会被赋值
         self.model_setting = self.backend_settings.models[self.model]
@@ -792,7 +800,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         else:
             tools_params = {}
 
-        if max_tokens is None:
+        if not max_tokens and not max_completion_tokens:
             max_output_tokens = self.model_setting.max_output_tokens
             token_counts = get_message_token_counts(messages=messages, tools=tools, model=self.model)
             if max_output_tokens is not None:
@@ -801,6 +809,10 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
             else:
                 max_tokens = self.model_setting.context_length - token_counts - 64
 
+        if "o3-mini" in self.model_id:
+            max_completion_tokens = max_tokens
+            max_tokens = NOT_GIVEN
+
         await self._acquire_rate_limit(self.endpoint, self.model, messages)
 
         if self.stream:

{vectorvein-0.2.67.dist-info → vectorvein-0.2.69.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.2.67
+Version: 0.2.69
 Summary: VectorVein Python SDK
 Author-Email: Anderson <andersonby@163.com>
 License: MIT

{vectorvein-0.2.67.dist-info → vectorvein-0.2.69.dist-info}/RECORD

@@ -1,13 +1,13 @@
-vectorvein-0.2.67.dist-info/METADATA,sha256=qWWjTXDYBxtywqJT-x7olVtPDI3rkFcW1CdwtMWBUoc,4567
-vectorvein-0.2.67.dist-info/WHEEL,sha256=thaaA2w1JzcGC48WYufAs8nrYZjJm8LqNfnXFOFyCC4,90
-vectorvein-0.2.67.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
+vectorvein-0.2.69.dist-info/METADATA,sha256=ICBJqenymd9STyy9YjP3-4xKUfYUikSdVorW-N8Uro0,4567
+vectorvein-0.2.69.dist-info/WHEEL,sha256=thaaA2w1JzcGC48WYufAs8nrYZjJm8LqNfnXFOFyCC4,90
+vectorvein-0.2.69.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=UIytpIgwo8qkZpIyrHVxLYTyliUOTp4J7C4iHRjbtWE,23850
-vectorvein/chat_clients/anthropic_client.py,sha256=yrGb87h_hpxhK-O3Ir0L5yMHLimPgzaJqdW4AjzskHk,68924
+vectorvein/chat_clients/anthropic_client.py,sha256=IjQOFq7Ifxlb5_nN9hXelE-Odf3zvYnQ3VWgzIkk5ZI,68976
 vectorvein/chat_clients/baichuan_client.py,sha256=CVMvpgjdrZGv0BWnTOBD-f2ufZ3wq3496wqukumsAr4,526
 vectorvein/chat_clients/base_client.py,sha256=p7s-G4Wh9MSpDKEfG8wuFAeWy5DGvj5Go31hqrpQPhM,38817
 vectorvein/chat_clients/deepseek_client.py,sha256=3qWu01NlJAP2N-Ff62d5-CZXZitlizE1fzb20LNetig,526
@@ -19,7 +19,7 @@ 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=3MEyKqlf1BtzbNpmqAHQq0JcYwlWxEzM0pOn0XNauas,47999
+vectorvein/chat_clients/openai_compatible_client.py,sha256=pKvUvZMcE2b2WAftF_XyO1m5WlUlvB5VBtvSGVW29H0,48613
 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
@@ -64,4 +64,4 @@ vectorvein/workflow/nodes/web_crawlers.py,sha256=FB0bTimkk___p3Z5UwQx2YarJyQCc45
 vectorvein/workflow/utils/check.py,sha256=Oj_S5WQf4_Fr_ro3ipjZt9unKFSFcuwZrrSmrS9kVLE,10193
 vectorvein/workflow/utils/json_to_code.py,sha256=ozdENkT1fYO67XdZX3Una5_jhGEUJdwD0DfZh0x4q3w,7120
 vectorvein/workflow/utils/layout.py,sha256=j0bRD3uaXu40xCS6U6BGahBI8FrHa5MiF55GbTrZ1LM,4565
-vectorvein-0.2.67.dist-info/RECORD,,
+vectorvein-0.2.69.dist-info/RECORD,,