vectorvein 0.1.9__py3-none-any.whl → 0.1.11__py3-none-any.whl

vectorvein/chat_clients/anthropic_client.py

@@ -118,6 +118,7 @@ class AnthropicChatClient(BaseChatClient):
         max_tokens: int | None = None,
         tools: list | NotGiven = NOT_GIVEN,
         tool_choice: str | NotGiven = NOT_GIVEN,
+        **kwargs,
     ):
         if model is not None:
             self.model = model
@@ -210,6 +211,7 @@ class AnthropicChatClient(BaseChatClient):
             max_tokens=max_tokens,
             tools=tools_params,
             tool_choice=tool_choice,
+            **kwargs,
         )
 
         if self.stream:
@@ -318,6 +320,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
         max_tokens: int | None = None,
         tools: list | NotGiven = NOT_GIVEN,
         tool_choice: str | NotGiven = NOT_GIVEN,
+        **kwargs,
     ):
         if model is not None:
             self.model = model
@@ -410,6 +413,7 @@ class AsyncAnthropicChatClient(BaseAsyncChatClient):
             max_tokens=max_tokens,
             tools=tools_params,
             tool_choice=tool_choice,
+            **kwargs,
         )
 
         if self.stream:

vectorvein/chat_clients/gemini_client.py

@@ -146,9 +146,9 @@ class GeminiChatClient(BaseChatClient):
             result = {
                 "content": "",
                 "usage": {
-                    "prompt_tokens": response["usageMetadata"]["promptTokenCount"],
-                    "completion_tokens": response["usageMetadata"]["candidatesTokenCount"],
-                    "total_tokens": response["usageMetadata"]["totalTokenCount"],
+                    "prompt_tokens": response.get("usageMetadata", {}).get("promptTokenCount", 0),
+                    "completion_tokens": response.get("usageMetadata", {}).get("candidatesTokenCount", 0),
+                    "total_tokens": response.get("usageMetadata", {}).get("totalTokenCount", 0),
                 },
             }
             tool_calls = []
@@ -309,9 +309,9 @@ class AsyncGeminiChatClient(BaseAsyncChatClient):
                 result = {
                     "content": "",
                     "usage": {
-                        "prompt_tokens": response["usageMetadata"]["promptTokenCount"],
-                        "completion_tokens": response["usageMetadata"]["candidatesTokenCount"],
-                        "total_tokens": response["usageMetadata"]["totalTokenCount"],
+                        "prompt_tokens": response.get("usageMetadata", {}).get("promptTokenCount", 0),
+                        "completion_tokens": response.get("usageMetadata", {}).get("candidatesTokenCount", 0),
+                        "total_tokens": response.get("usageMetadata", {}).get("totalTokenCount", 0),
                     },
                 }
                 tool_calls = []

vectorvein/chat_clients/minimax_client.py

@@ -25,7 +25,7 @@ def extract_tool_calls(response):
                         "index": index,
                         "id": tool_call["id"],
                         "function": tool_call["function"],
-                        "type": tool_call["type"],
+                        "type": "function",
                     }
                     for index, tool_call in enumerate(tool_calls)
                 ]
@@ -59,6 +59,7 @@ class MiniMaxChatClient(BaseChatClient):
             endpoint_id,
             **kwargs,
         )
+        self.http_client = httpx.Client()
 
     def create_completion(
         self,
@@ -69,6 +70,7 @@ class MiniMaxChatClient(BaseChatClient):
         max_tokens: int | None = None,
         tools: list | None = None,
         tool_choice: str = "auto",
+        **kwargs,
     ):
         if model is not None:
             self.model = model
@@ -134,43 +136,50 @@ class MiniMaxChatClient(BaseChatClient):
             "stream": self.stream,
             "mask_sensitive_info": False,
             **tools_params,
+            **kwargs,
         }
 
-        response = httpx.post(
-            url=self.url,
-            headers=self.headers,
-            json=request_body,
-            timeout=60,
-        )
-
         if self.stream:
 
             def generator():
-                for chunk in response.iter_lines():
-                    if chunk:
-                        chunk_data = json.loads(chunk[6:])
-                        if chunk_data["object"] != "chat.completion.chunk":
-                            continue
-                        tool_calls_params = extract_tool_calls(chunk_data)
-                        has_tool_calls = True if tool_calls_params else False
-                        if has_tool_calls:
-                            yield ChatCompletionDeltaMessage(
-                                **{
-                                    "content": chunk_data["choices"][0]["delta"].get("content"),
-                                    "role": "assistant",
-                                    **tool_calls_params,
-                                }
-                            )
-                        else:
-                            yield ChatCompletionDeltaMessage(
-                                **{
-                                    "content": chunk_data["choices"][0]["delta"]["content"],
-                                    "role": "assistant",
-                                }
-                            )
+                with self.http_client.stream(
+                    "POST",
+                    url=self.url,
+                    headers=self.headers,
+                    json=request_body,
+                    timeout=60,
+                ) as response:
+                    for chunk in response.iter_lines():
+                        if chunk:
+                            chunk_data = json.loads(chunk[6:])
+                            if chunk_data["object"] != "chat.completion.chunk":
+                                continue
+                            tool_calls_params = extract_tool_calls(chunk_data)
+                            has_tool_calls = True if tool_calls_params else False
+                            if has_tool_calls:
+                                yield ChatCompletionDeltaMessage(
+                                    **{
+                                        "content": chunk_data["choices"][0]["delta"].get("content"),
+                                        "role": "assistant",
+                                        **tool_calls_params,
+                                    }
+                                )
+                            else:
+                                yield ChatCompletionDeltaMessage(
+                                    **{
+                                        "content": chunk_data["choices"][0]["delta"]["content"],
+                                        "role": "assistant",
+                                    }
+                                )
 
             return generator()
         else:
+            response = httpx.post(
+                url=self.url,
+                headers=self.headers,
+                json=request_body,
+                timeout=60,
+            )
             result = response.json()
             tool_calls_params = extract_tool_calls(result)
             return ChatCompletionMessage(
@@ -221,6 +230,7 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
         max_tokens: int | None = None,
         tools: list | None = None,
         tool_choice: str = "auto",
+        **kwargs,
     ):
         if model is not None:
             self.model = model
@@ -284,6 +294,7 @@ class AsyncMiniMaxChatClient(BaseAsyncChatClient):
             "stream": self.stream,
             "mask_sensitive_info": False,
             **tools_params,
+            **kwargs,
         }
 
         if self.stream:

vectorvein/chat_clients/openai_compatible_client.py

@@ -54,6 +54,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
         max_tokens: int | None = None,
         tools: list | NotGiven = NOT_GIVEN,
         tool_choice: str | NotGiven = NOT_GIVEN,
+        **kwargs,
     ):
         if model is not None:
             self.model = model
@@ -120,6 +121,7 @@ class OpenAICompatibleChatClient(BaseChatClient):
             temperature=self.temperature,
             max_tokens=max_tokens,
             **tools_params,
+            **kwargs,
         )
 
         if self.stream:
@@ -207,6 +209,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
         max_tokens: int | None = None,
         tools: list | NotGiven = NOT_GIVEN,
         tool_choice: str | NotGiven = NOT_GIVEN,
+        **kwargs,
     ):
         if model is not None:
             self.model = model
@@ -273,6 +276,7 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
             temperature=self.temperature,
             max_tokens=max_tokens,
             **tools_params,
+            **kwargs,
         )
 
         if self.stream:

vectorvein/types/defaults.py

@@ -213,6 +213,13 @@ ZHIPUAI_MODELS = {
         "response_format_available": False,
         "max_output_tokens": 4095,
     },
+    "glm-4-plus": {
+        "id": "glm-4-plus",
+        "context_length": 128000,
+        "function_call_available": True,
+        "response_format_available": False,
+        "max_output_tokens": 4095,
+    },
     "glm-4-0520": {
         "id": "glm-4-0520",
         "context_length": 128000,
@@ -255,6 +262,13 @@ ZHIPUAI_MODELS = {
         "response_format_available": False,
         "max_output_tokens": 1024,
     },
+    "glm-4v-plus": {
+        "id": "glm-4v-plus",
+        "context_length": 2000,
+        "function_call_available": False,
+        "response_format_available": False,
+        "max_output_tokens": 1024,
+    },
 }
 
 # Mistral models

{vectorvein-0.1.9.dist-info → vectorvein-0.1.11.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.1.9
+Version: 0.1.11
 Summary: Default template for PDM package
 Author-Email: Anderson <andersonby@163.com>
 License: MIT

{vectorvein-0.1.9.dist-info → vectorvein-0.1.11.dist-info}/RECORD

@@ -1,26 +1,26 @@
-vectorvein-0.1.9.dist-info/METADATA,sha256=AlikMRU7DLdZ6gZMohsL1X6NiuAWP3jGV0tE-uZkhNo,501
-vectorvein-0.1.9.dist-info/WHEEL,sha256=rSwsxJWe3vzyR5HCwjWXQruDgschpei4h_giTm0dJVE,90
+vectorvein-0.1.11.dist-info/METADATA,sha256=rw0fFY-7V-teJUIt4t4KlvLIZHp4hjgaQBRpQ816eP0,502
+vectorvein-0.1.11.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=JjigSUsIn06ixIEjnOJhVbcMqy2_MAL3iVUlDFAFMW4,20008
+vectorvein/chat_clients/anthropic_client.py,sha256=MHpLywbIIgFsAkj8EiKEoxUgxcbx3njFxLTN2GhEjys,20088
 vectorvein/chat_clients/base_client.py,sha256=wMXpQ1L1KDb2Hg6va3H3GmcVeQB6r6sh7F4IS0DBQWI,4275
 vectorvein/chat_clients/deepseek_client.py,sha256=3qWu01NlJAP2N-Ff62d5-CZXZitlizE1fzb20LNetig,526
-vectorvein/chat_clients/gemini_client.py,sha256=IHcBHTSHkj3f962S5L7Ga-XA-96sq8quIDRZpoqvGss,13653
+vectorvein/chat_clients/gemini_client.py,sha256=jUu-wwpaChP8_lxLdWnYkgVr_aSegCHbmBgnpFztqAw,13743
 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=uomp3DyTBmDXQtCmRiYp1VIIOFoVZ9_oyM3-j4JO7go,13000
+vectorvein/chat_clients/minimax_client.py,sha256=ZO-7WNsvsct3cpuo0Qc6WL3A0kkijn0J8o0ahlxmUyE,13478
 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=fvg--wFwnFEEhLGS9_u1XzNhtkkDUf4_rq6zYKwnOuI,13738
+vectorvein/chat_clients/openai_compatible_client.py,sha256=rPxIPj4STm1sy7T6bqyZpO-FPwMyjyQmThyJqd8_jYs,13822
 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
 vectorvein/chat_clients/zhipuai_client.py,sha256=Ys5DSeLCuedaDXr3PfG1EW2zKXopt-awO2IylWSwY0s,519
 vectorvein/settings/__init__.py,sha256=4mpccT7eZC3yI1vVnVViW4wHBnDEH9D2R5EsIP34VgU,3218
-vectorvein/types/defaults.py,sha256=ANIYL0W0bxl2IBxvtkS_WlS_qMQQwpi5TKRdLxdk47M,13027
+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/utilities/media_processing.py,sha256=BujciRmw1GMmc3ELRvafL8STcy6r5b2rVnh27-uA7so,2256
 vectorvein/utilities/retry.py,sha256=9ePuJdeUUGx-qMWfaFxmlOvG_lQPwCQ4UB1z3Edlo34,993
-vectorvein-0.1.9.dist-info/RECORD,,
+vectorvein-0.1.11.dist-info/RECORD,,