vectorvein 0.1.46__py3-none-any.whl → 0.1.48__py3-none-any.whl

vectorvein/chat_clients/base_client.py

@@ -148,6 +148,7 @@ class BaseChatClient(ABC):
             tools=tools,
             tool_choice=tool_choice,
             response_format=response_format,
+            stream_options=stream_options,
             **kwargs,
         )
 

vectorvein/chat_clients/openai_compatible_client.py

@@ -6,8 +6,6 @@ from functools import cached_property
 from typing import overload, Generator, AsyncGenerator, Any, Literal, Iterable
 
 import httpx
-from openai._streaming import Stream, AsyncStream
-from openai.types.chat import ChatCompletion, ChatCompletionChunk
 from openai import OpenAI, AsyncOpenAI, AzureOpenAI, AsyncAzureOpenAI
 
 from .base_client import BaseChatClient, BaseAsyncChatClient
@@ -202,7 +200,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
             _stream_options_params = {}
 
         if self.stream:
-            stream_response: Stream[ChatCompletionChunk] = raw_client.chat.completions.create(
+            stream_response = raw_client.chat.completions.create(
                 model=self.model_id,
                 messages=messages,
                 stream=True,
@@ -225,11 +223,14 @@ class OpenAICompatibleChatClient(BaseChatClient):
                             prompt_tokens=chunk.usage.prompt_tokens or 0,
                             total_tokens=chunk.usage.total_tokens or 0,
                         )
-                    else:
-                        usage = None
+
                     if len(chunk.choices) == 0:
+                        if usage:
+                            yield ChatCompletionDeltaMessage(usage=usage)
                         continue
                     if not chunk.choices[0].delta:
+                        if usage:
+                            yield ChatCompletionDeltaMessage(usage=usage)
                         continue
                     if self.model_setting.function_call_available:
                         if chunk.choices[0].delta.tool_calls:
@@ -253,7 +254,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
 
             return generator()
         else:
-            response: ChatCompletion = raw_client.chat.completions.create(
+            response = raw_client.chat.completions.create(
                 model=self.model_id,
                 messages=messages,
                 stream=False,
@@ -453,7 +454,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
                 max_tokens = self.model_setting.context_length - token_counts - 64
 
         if self.stream:
-            stream_response: AsyncStream[ChatCompletionChunk] = await raw_client.chat.completions.create(
+            stream_response = await raw_client.chat.completions.create(
                 model=self.model_id,
                 messages=messages,
                 stream=self.stream,
@@ -476,12 +477,16 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
                             prompt_tokens=chunk.usage.prompt_tokens or 0,
                             total_tokens=chunk.usage.total_tokens or 0,
                         )
-                    else:
-                        usage = None
+
                     if len(chunk.choices) == 0:
+                        if usage:
+                            yield ChatCompletionDeltaMessage(usage=usage)
                         continue
                     if not chunk.choices[0].delta:
+                        if usage:
+                            yield ChatCompletionDeltaMessage(usage=usage)
                         continue
+
                     if self.model_setting.function_call_available:
                         if chunk.choices[0].delta.tool_calls:
                             for index, tool_call in enumerate(chunk.choices[0].delta.tool_calls):
@@ -504,7 +509,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
 
             return generator()
         else:
-            response: ChatCompletion = await raw_client.chat.completions.create(
+            response = await raw_client.chat.completions.create(
                 model=self.model_id,
                 messages=messages,
                 stream=self.stream,

{vectorvein-0.1.46.dist-info → vectorvein-0.1.48.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.1.46
+Version: 0.1.48
 Summary: Default template for PDM package
 Author-Email: Anderson <andersonby@163.com>
 License: MIT

{vectorvein-0.1.46.dist-info → vectorvein-0.1.48.dist-info}/RECORD

@@ -1,11 +1,11 @@
-vectorvein-0.1.46.dist-info/METADATA,sha256=36D8mKjmVS10vBTLPCUu1sMLJoJQWV4SQS_AQeFJRN4,644
-vectorvein-0.1.46.dist-info/WHEEL,sha256=thaaA2w1JzcGC48WYufAs8nrYZjJm8LqNfnXFOFyCC4,90
-vectorvein-0.1.46.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
+vectorvein-0.1.48.dist-info/METADATA,sha256=J6bEBBrKiwUMNvIysDT5W-V8fDFhdwIYPjWBYXmFRKM,644
+vectorvein-0.1.48.dist-info/WHEEL,sha256=thaaA2w1JzcGC48WYufAs8nrYZjJm8LqNfnXFOFyCC4,90
+vectorvein-0.1.48.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
 vectorvein/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 vectorvein/chat_clients/__init__.py,sha256=dW169oK1n3v8Z0uD8itghzlCP72rxiaS-XYn6fvI2xM,16788
 vectorvein/chat_clients/anthropic_client.py,sha256=i1fMYSkUovd-Lc9B64bWMgzSTUFw4S3fj3AJ_pDokT4,34029
 vectorvein/chat_clients/baichuan_client.py,sha256=CVMvpgjdrZGv0BWnTOBD-f2ufZ3wq3496wqukumsAr4,526
-vectorvein/chat_clients/base_client.py,sha256=4IYRAfRt9jk1nuUh7e57boaI7_WHR31fJKvvRkt3RDE,9454
+vectorvein/chat_clients/base_client.py,sha256=0Uj0e-JR0a68sRS_WfUMVd91Av7lzJh6-DukjutlaD0,9497
 vectorvein/chat_clients/deepseek_client.py,sha256=3qWu01NlJAP2N-Ff62d5-CZXZitlizE1fzb20LNetig,526
 vectorvein/chat_clients/gemini_client.py,sha256=TJZUP8dbufuDnqkRI_gxIDDCYHrIhMPCqKGI6sSwqZo,19860
 vectorvein/chat_clients/groq_client.py,sha256=Uow4pgdmFi93ZQSoOol2-0PhhqkW-S0XuSldvppz5U4,498
@@ -14,7 +14,7 @@ vectorvein/chat_clients/minimax_client.py,sha256=cLpSJ5vIaDi0Kmrt7XGlbjnlFnByJiI
 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=CUylVzn-1bYB27gDDUVtfjxeSqG7yirFssNFW2dpBaQ,21941
+vectorvein/chat_clients/openai_compatible_client.py,sha256=D2VmhpDVct4w2y58s87An7x22koOdkxSOKR2-so9lJI,22044
 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
@@ -32,4 +32,4 @@ vectorvein/types/llm_parameters.py,sha256=5o-C_yXxxQWZy_e8OWowB2107GTS-Eawx4Mvb1
 vectorvein/types/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 vectorvein/utilities/media_processing.py,sha256=cnzLrU1OaJvSv87IOnc36FrDXtmGMDStPbxtIJ33YN4,5880
 vectorvein/utilities/retry.py,sha256=6KFS9R2HdhqM3_9jkjD4F36ZSpEx2YNFGOVlpOsUetM,2208
-vectorvein-0.1.46.dist-info/RECORD,,
+vectorvein-0.1.48.dist-info/RECORD,,