gllm-inference-binary 0.5.8__cp311-cp311-macosx_13_0_x86_64.whl → 0.5.10__cp311-cp311-macosx_13_0_x86_64.whl

gllm_inference/constants.pyi

@@ -7,3 +7,4 @@ INVOKER_PROPAGATED_MAX_RETRIES: int
 INVOKER_DEFAULT_TIMEOUT: float
 HEX_REPR_LENGTH: int
 HTTP_STATUS_CODE_PATTERNS: Incomplete
+SECONDS_TO_MILLISECONDS: int

gllm_inference/em_invoker/google_em_invoker.pyi

@@ -1,6 +1,6 @@
 from _typeshed import Incomplete
 from gllm_core.utils.retry import RetryConfig as RetryConfig
-from gllm_inference.constants import GOOGLE_SCOPES as GOOGLE_SCOPES
+from gllm_inference.constants import GOOGLE_SCOPES as GOOGLE_SCOPES, SECONDS_TO_MILLISECONDS as SECONDS_TO_MILLISECONDS
 from gllm_inference.em_invoker.em_invoker import BaseEMInvoker as BaseEMInvoker
 from gllm_inference.em_invoker.schema.google import Key as Key
 from gllm_inference.schema import ModelId as ModelId, ModelProvider as ModelProvider, Vector as Vector

gllm_inference/lm_invoker/google_lm_invoker.pyi

@@ -2,7 +2,7 @@ from _typeshed import Incomplete
 from gllm_core.event import EventEmitter as EventEmitter
 from gllm_core.schema.tool import Tool
 from gllm_core.utils.retry import RetryConfig as RetryConfig
-from gllm_inference.constants import GOOGLE_SCOPES as GOOGLE_SCOPES
+from gllm_inference.constants import GOOGLE_SCOPES as GOOGLE_SCOPES, SECONDS_TO_MILLISECONDS as SECONDS_TO_MILLISECONDS
 from gllm_inference.lm_invoker.lm_invoker import BaseLMInvoker as BaseLMInvoker
 from gllm_inference.lm_invoker.schema.google import InputType as InputType, Key as Key
 from gllm_inference.schema import Attachment as Attachment, AttachmentType as AttachmentType, EmitDataType as EmitDataType, LMOutput as LMOutput, Message as Message, MessageRole as MessageRole, ModelId as ModelId, ModelProvider as ModelProvider, Reasoning as Reasoning, ResponseSchema as ResponseSchema, TokenUsage as TokenUsage, ToolCall as ToolCall, ToolResult as ToolResult
@@ -162,7 +162,12 @@ class GoogleLMInvoker(BaseLMInvoker):
         ```python
         LMOutput(
             response="Golden retriever is a good dog breed.",
-            token_usage=TokenUsage(input_tokens=100, output_tokens=50),
+            token_usage=TokenUsage(
+                input_tokens=1500,
+                output_tokens=200,
+                input_token_details=InputTokenDetails(cached_tokens=1200, uncached_tokens=300),
+                output_token_details=OutputTokenDetails(reasoning_tokens=180, response_tokens=20),
+            ),
             duration=0.729,
             finish_details={"finish_reason": "STOP", "finish_message": None},
         )

gllm_inference/schema/__init__.pyi

@@ -5,9 +5,9 @@ from gllm_inference.schema.lm_output import LMOutput as LMOutput
 from gllm_inference.schema.message import Message as Message
 from gllm_inference.schema.model_id import ModelId as ModelId, ModelProvider as ModelProvider
 from gllm_inference.schema.reasoning import Reasoning as Reasoning
-from gllm_inference.schema.token_usage import TokenUsage as TokenUsage
+from gllm_inference.schema.token_usage import InputTokenDetails as InputTokenDetails, OutputTokenDetails as OutputTokenDetails, TokenUsage as TokenUsage
 from gllm_inference.schema.tool_call import ToolCall as ToolCall
 from gllm_inference.schema.tool_result import ToolResult as ToolResult
 from gllm_inference.schema.type_alias import EMContent as EMContent, ErrorResponse as ErrorResponse, MessageContent as MessageContent, ResponseSchema as ResponseSchema, Vector as Vector
 
-__all__ = ['Attachment', 'AttachmentType', 'CodeExecResult', 'EMContent', 'EmitDataType', 'ErrorResponse', 'MessageContent', 'LMOutput', 'ModelId', 'ModelProvider', 'Message', 'MessageRole', 'Reasoning', 'ResponseSchema', 'TokenUsage', 'ToolCall', 'ToolResult', 'Vector']
+__all__ = ['Attachment', 'AttachmentType', 'CodeExecResult', 'EMContent', 'EmitDataType', 'ErrorResponse', 'InputTokenDetails', 'MessageContent', 'LMOutput', 'ModelId', 'ModelProvider', 'Message', 'MessageRole', 'OutputTokenDetails', 'Reasoning', 'ResponseSchema', 'TokenUsage', 'ToolCall', 'ToolResult', 'Vector']

gllm_inference/schema/token_usage.pyi

@@ -1,11 +1,75 @@
 from pydantic import BaseModel
 
+class InputTokenDetails(BaseModel):
+    """Defines the input token details schema.
+
+    Attributes:
+        cached_tokens (int): The number of cached tokens. Defaults to 0.
+        uncached_tokens (int): The number of uncached tokens. Defaults to 0.
+    """
+    cached_tokens: int
+    uncached_tokens: int
+    def __add__(self, other: InputTokenDetails) -> InputTokenDetails:
+        """Add two InputTokenDetails objects together.
+
+        Args:
+            other (InputTokenDetails): The other InputTokenDetails object to add.
+
+        Returns:
+            InputTokenDetails: A new InputTokenDetails object with summed values.
+        """
+
+class OutputTokenDetails(BaseModel):
+    """Defines the output token details schema.
+
+    Attributes:
+        reasoning_tokens (int): The number of reasoning tokens. Defaults to 0.
+        response_tokens (int): The number of response tokens. Defaults to 0.
+    """
+    reasoning_tokens: int
+    response_tokens: int
+    def __add__(self, other: OutputTokenDetails) -> OutputTokenDetails:
+        """Add two OutputTokenDetails objects together.
+
+        Args:
+            other (OutputTokenDetails): The other OutputTokenDetails object to add.
+
+        Returns:
+            OutputTokenDetails: A new OutputTokenDetails object with summed values.
+        """
+
 class TokenUsage(BaseModel):
     """Defines the token usage data structure of a language model.
 
     Attributes:
-        input_tokens (int): The number of input tokens.
-        output_tokens (int): The number of output tokens.
+        input_tokens (int): The number of input tokens. Defaults to 0.
+        output_tokens (int): The number of output tokens. Defaults to 0.
+        input_token_details (InputTokenDetails | None): The details of the input tokens. Defaults to None.
+        output_token_details (OutputTokenDetails | None): The details of the output tokens. Defaults to None.
     """
     input_tokens: int
     output_tokens: int
+    input_token_details: InputTokenDetails | None
+    output_token_details: OutputTokenDetails | None
+    @classmethod
+    def from_token_details(cls, input_tokens: int | None = None, output_tokens: int | None = None, cached_tokens: int | None = None, reasoning_tokens: int | None = None) -> TokenUsage:
+        """Creates a TokenUsage from token details.
+
+        Args:
+            input_tokens (int | None): The number of input tokens. Defaults to None.
+            output_tokens (int | None): The number of output tokens. Defaults to None.
+            cached_tokens (int | None): The number of cached tokens. Defaults to None.
+            reasoning_tokens (int | None): The number of reasoning tokens. Defaults to None.
+
+        Returns:
+            TokenUsage: The instantiated TokenUsage.
+        """
+    def __add__(self, other: TokenUsage) -> TokenUsage:
+        """Add two TokenUsage objects together.
+
+        Args:
+            other (TokenUsage): The other TokenUsage object to add.
+
+        Returns:
+            TokenUsage: A new TokenUsage object with summed values.
+        """

{gllm_inference_binary-0.5.8.dist-info → gllm_inference_binary-0.5.10.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gllm-inference-binary
-Version: 0.5.8
+Version: 0.5.10
 Summary: A library containing components related to model inferences in Gen AI applications.
 Author: Henry Wicaksono
 Author-email: henry.wicaksono@gdplabs.id

{gllm_inference_binary-0.5.8.dist-info → gllm_inference_binary-0.5.10.dist-info}/RECORD

@@ -8,11 +8,11 @@ gllm_inference/catalog/__init__.pyi,sha256=JBkPGTyiiZ30GECzJBW-mW8LekWyY2qyzal3e
 gllm_inference/catalog/catalog.pyi,sha256=a4RNG1lKv51GxQpOqh47tz-PAROMPaeP2o5XNLBSZaU,4790
 gllm_inference/catalog/lm_request_processor_catalog.pyi,sha256=ranHMbG9--DZj9FJRhIUa6U8e-L-Tm-_hSBpzJ6DDs4,5428
 gllm_inference/catalog/prompt_builder_catalog.pyi,sha256=OU8k_4HbqjZEzHZlzSM3uzGQZJmM2uGD76Csqom0CEQ,3197
-gllm_inference/constants.pyi,sha256=gPlwRHKIjUdyQoLdoog8ca76YmjrQL0SCDTEn8UEWSY,256
+gllm_inference/constants.pyi,sha256=JDSc_GdTGg2RVlwM9Jy-2dtC4a5311KcE1WHdnUMmt4,285
 gllm_inference/em_invoker/__init__.pyi,sha256=XESsrYo1PZeeHe7AMRyuzKoV7XDD5oN89ZTH01zRf4k,873
 gllm_inference/em_invoker/azure_openai_em_invoker.pyi,sha256=OEkVu5nv92ITqdhDtgDg4MiLSDRWDmLSnAhYtXpCn6E,4602
 gllm_inference/em_invoker/em_invoker.pyi,sha256=hiH8FB5R-KxhI8Ds2htF3cjRcIcH92yHPcOdpgc4FDo,4341
-gllm_inference/em_invoker/google_em_invoker.pyi,sha256=pn05VdOZXZOffeg89wlAQOBAyHyt6dLaS0Sd6LoH05M,6130
+gllm_inference/em_invoker/google_em_invoker.pyi,sha256=vj0KAX5b2rhvqqbpjFZYLmk84RHuNw0pXkhx3bbHmM4,6182
 gllm_inference/em_invoker/langchain/__init__.pyi,sha256=VYGKE5OgU0my1RlhgzkU_A7-GLGnUDDnNFuctuRwILE,148
 gllm_inference/em_invoker/langchain/em_invoker_embeddings.pyi,sha256=6nASLqi0FXCpqyYPl7kM3g7hAW-xS5ZwsS3GFudns98,2347
 gllm_inference/em_invoker/langchain_em_invoker.pyi,sha256=lXiTTGcNOIwurZx3_6vWLE1DQyioK8Z1fOcPuUATnxA,2782
@@ -35,7 +35,7 @@ gllm_inference/lm_invoker/anthropic_lm_invoker.pyi,sha256=MsF3OmDo0L9aEHuTJYTgso
 gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi,sha256=EV_yrj6mnV_rCDEqSZaIikfw76_rXXyDlC-w_y-m7K0,14603
 gllm_inference/lm_invoker/bedrock_lm_invoker.pyi,sha256=IuLxgCThOSBHx1AXqnhL6yVu5_JV6hAeGBWWm5P1JCo,12423
 gllm_inference/lm_invoker/datasaur_lm_invoker.pyi,sha256=QS84w3WpD3Oyl5HdxrucsadCmsHE8gn6Ewl3l01DCgI,9203
-gllm_inference/lm_invoker/google_lm_invoker.pyi,sha256=Hy5DYaIaKPqMjSjDyfHy0HofHhogAn3McN7SeDs67g4,16491
+gllm_inference/lm_invoker/google_lm_invoker.pyi,sha256=aPOlaw2rexUDhbMFaXnuKqOT7lqeKxjfeToe9LjwEUw,16787
 gllm_inference/lm_invoker/langchain_lm_invoker.pyi,sha256=skcbX34ZosGMFli3SLmGITJTzMj9HxkQBWGrd-zYCbU,13264
 gllm_inference/lm_invoker/litellm_lm_invoker.pyi,sha256=_c56ewpEQ-Ywj5ofFzRYBvQgefR7Q_WkcQt97lnIFgg,13128
 gllm_inference/lm_invoker/lm_invoker.pyi,sha256=NLLzJY5YaWckR0DHgNNxZhg-Lif41ZX-DZZcICq7NIU,7809
@@ -74,7 +74,7 @@ gllm_inference/prompt_formatter/prompt_formatter.pyi,sha256=UkcPi5ao98OGJyNRsqfh
 gllm_inference/request_processor/__init__.pyi,sha256=hVnfdNZnkTBJHnmLtN3Na4ANP0yK6AstWdIizVr2Apo,227
 gllm_inference/request_processor/lm_request_processor.pyi,sha256=7pVNb2GwITb1jTflZP498qZ321G15b16jayZuuhuO1o,5424
 gllm_inference/request_processor/uses_lm_mixin.pyi,sha256=He-ytjwv2H5Hn312WFBAlBK96ALKTtDO3AT_80hCGTg,2321
-gllm_inference/schema/__init__.pyi,sha256=bJeO_4fCFPmYrVZ-4RQw0eQn62X228pB0w2tYDM-oek,1238
+gllm_inference/schema/__init__.pyi,sha256=aQx7eKVsFNgJkYtbbKppa9Ww2WfTe0yzl_ewajDCo6w,1363
 gllm_inference/schema/attachment.pyi,sha256=jApuzjOHJDCz4lr4MlHzBgIndh559nbWu2Xp1fk3hso,3297
 gllm_inference/schema/code_exec_result.pyi,sha256=ZTHh6JtRrPIdQ059P1UAiD2L-tAO1_S5YcMsAXfJ5A0,559
 gllm_inference/schema/enums.pyi,sha256=Nvc_Qsd8yyiH_tCtymN39O6EZ0DT_wxYHmlKLllxC00,605
@@ -82,7 +82,7 @@ gllm_inference/schema/lm_output.pyi,sha256=GafJV0KeD-VSwWkwG1oz-uruXrQ7KDZTuoojP
 gllm_inference/schema/message.pyi,sha256=VP9YppKj2mo1esl9cy6qQO9m2mMHUjTmfGDdyUor880,2220
 gllm_inference/schema/model_id.pyi,sha256=XFfIkfetFamwVxmUifMZLehuvTcASkK7kp4OMNcteo8,5230
 gllm_inference/schema/reasoning.pyi,sha256=SlTuiDw87GdnAn-I6YOPIJRhEBiwQljM46JohG05guQ,562
-gllm_inference/schema/token_usage.pyi,sha256=fKXfdd-q6HrP0h7zzj16OJ5WEOn2YedsCFIgd5IyDFk,312
+gllm_inference/schema/token_usage.pyi,sha256=1GTQVORV0dBNmD_jix8aVaUqxMKFF04KpLP7y2urqbk,2950
 gllm_inference/schema/tool_call.pyi,sha256=zQaVxCnkVxOfOEhBidqohU85gb4PRwnwBiygKaunamk,389
 gllm_inference/schema/tool_result.pyi,sha256=cAG7TVtB4IWJPt8XBBbB92cuY1ZsX9M276bN9aqjcvM,276
 gllm_inference/schema/type_alias.pyi,sha256=CkqX5zLML4vII7BEIXDz7ZQd211RsHtq7EJekkV2V6g,725
@@ -90,8 +90,8 @@ gllm_inference/utils/__init__.pyi,sha256=npmBmmlBv7cPHMg1hdL3S2_RelD6vk_LhCsGELh
 gllm_inference/utils/langchain.pyi,sha256=VluQiHkGigDdqLUbhB6vnXiISCP5hHqV0qokYY6dC1A,1164
 gllm_inference/utils/validation.pyi,sha256=toxBtRp-VItC_X7sNi-GDd7sjibBdWMrR0q01OI2D7k,385
 gllm_inference.build/.gitignore,sha256=aEiIwOuxfzdCmLZe4oB1JsBmCUxwG8x-u-HBCV9JT8E,1
-gllm_inference.cpython-311-darwin.so,sha256=CU8q2Ky6SzAsZydA9FcY0mvBtO3RSG4tg-DLpLqpLTs,3814224
+gllm_inference.cpython-311-darwin.so,sha256=Up5mRt3VbXk-4X2piMNoF5vz_w2b3WJ9WT6p14t284U,3850688
 gllm_inference.pyi,sha256=fsNCXXsB4E8WhP477yGq_QOJAfOyoZA4G2PfAMBav5Y,3324
-gllm_inference_binary-0.5.8.dist-info/METADATA,sha256=czkvBDQEqltXhBF08Zq4Ik6KjTwuRDDvbafy9pXvs0Y,4531
-gllm_inference_binary-0.5.8.dist-info/WHEEL,sha256=r3EiIdyNg8wC0u2K9wWWq7Elb6S4XGGmkyBqljSOtNU,107
-gllm_inference_binary-0.5.8.dist-info/RECORD,,
+gllm_inference_binary-0.5.10.dist-info/METADATA,sha256=-9APgoHwt5Dnb10ZCbLWtABANZ_q1vYocrnW9kap5Bc,4532
+gllm_inference_binary-0.5.10.dist-info/WHEEL,sha256=r3EiIdyNg8wC0u2K9wWWq7Elb6S4XGGmkyBqljSOtNU,107
+gllm_inference_binary-0.5.10.dist-info/RECORD,,