glm-4.5 model

Author: jinhaiyang

zhipuai/api_resource/chat/async_completions.py

@@ -44,6 +44,8 @@ def create(
             extra_headers: Headers | None = None,
             extra_body: Body | None = None,
             timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+            response_format: object | None = None,
+            thinking: object | None = None,
     ) -> AsyncTaskStatus:
         _cast_type = AsyncTaskStatus
         logger.debug(f"temperature:{temperature}, top_p:{top_p}")
@@ -87,6 +89,8 @@ def create(
             "tool_choice": tool_choice,
             "meta": meta,
             "extra": maybe_transform(extra, code_geex_params.CodeGeexExtra),
+            "response_format": response_format,
+            "thinking": thinking
         }
         return self._post(
             "/async/chat/completions",

zhipuai/api_resource/chat/completions.py

@@ -47,7 +47,8 @@ def create(
             extra_headers: Headers | None = None,
             extra_body: Body | None = None,
             timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-            response_format : object | None = None
+            response_format : object | None = None,
+            thinking :  object | None = None,
     ) -> Completion | StreamResponse[ChatCompletionChunk]:
         logger.debug(f"temperature:{temperature}, top_p:{top_p}")
         if temperature is not None and temperature != NOT_GIVEN:
@@ -91,7 +92,8 @@ def create(
             "tool_choice": tool_choice,
             "meta": meta,
             "extra": maybe_transform(extra, code_geex_params.CodeGeexExtra),
-            "response_format": response_format
+            "response_format": response_format,
+            "thinking": thinking
         })
         return self._post(
             "/chat/completions",

zhipuai/types/chat/chat_completion_chunk.py

@@ -49,13 +49,19 @@ class Choice(BaseModel):
     finish_reason: Optional[str] = None
     index: int
 
+class PromptTokensDetails(BaseModel):
+    cached_tokens: int
+
+class CompletionTokensDetails(BaseModel):
+    reasoning_tokens: int
 
 class CompletionUsage(BaseModel):
     prompt_tokens: int
+    prompt_tokens_details: Optional[PromptTokensDetails] = None
     completion_tokens: int
+    completion_tokens_details: Optional[CompletionTokensDetails] = None
     total_tokens: int
 
-
 class ChatCompletionChunk(BaseModel):
     id: Optional[str] = None
     choices: List[Choice]