gllm-inference-binary 0.5.9__py3-none-any.whl → 0.5.10b3__py3-none-any.whl

gllm_inference/lm_invoker/google_lm_invoker.pyi

@@ -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-0.5.9.dist-info → gllm_inference_binary-0.5.10b3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gllm-inference-binary
-Version: 0.5.9
+Version: 0.5.10b3
 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-0.5.9.dist-info → gllm_inference_binary-0.5.10b3.dist-info}/RECORD

@@ -35,7 +35,7 @@ gllm_inference/lm_invoker/anthropic_lm_invoker.pyi,sha256=85uvShLv4-eiGOpTMgwWpQ
 gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi,sha256=N2TjGz5Gi6xiLkAgI6SzWq_V3tj66HJfMNff7d04uU0,14856
 gllm_inference/lm_invoker/bedrock_lm_invoker.pyi,sha256=ae5P_9sjtcOgMIUaRchvp8F0FujoeP4e2F_OoHSe_go,12655
 gllm_inference/lm_invoker/datasaur_lm_invoker.pyi,sha256=c4H3TOz0LIhWjokCCdQ4asiwQR4_LPyaimo4RAqU9es,9369
-gllm_inference/lm_invoker/google_lm_invoker.pyi,sha256=Ph9QmTdFlar92iUDDcWwwVY_KDkt9Rumfi41zpTNfRg,16853
+gllm_inference/lm_invoker/google_lm_invoker.pyi,sha256=I3plg_oVuTl0hiShFBmCYPclP4gWbzU61xUSgon24Ew,17102
 gllm_inference/lm_invoker/langchain_lm_invoker.pyi,sha256=bBGOxJfjnzOtDR4kH4PuCiOCKEPu8rTqzZodTXCHQ2k,13522
 gllm_inference/lm_invoker/litellm_lm_invoker.pyi,sha256=HHwW7i8ryXHI23JZQwscyva6aPmPOB13Muhf7gaaMUM,13376
 gllm_inference/lm_invoker/lm_invoker.pyi,sha256=YNJ0Sh_BOl1WbC69xvuxWM75qyByXJSXAYWSwtQ84cc,7960
@@ -74,7 +74,7 @@ gllm_inference/prompt_formatter/prompt_formatter.pyi,sha256=hAc6rxWc6JSYdD-OypLi
 gllm_inference/request_processor/__init__.pyi,sha256=giEme2WFQhgyKiBZHhSet0_nKSCHwGy-_2p6NRzg0Zc,231
 gllm_inference/request_processor/lm_request_processor.pyi,sha256=rInXhC95BvQnw9q98KZWpjPH8Q_TV4zC2ycNjypEBZ4,5516
 gllm_inference/request_processor/uses_lm_mixin.pyi,sha256=znBG4AWWm_H70Qqrc1mO4ohmWotX9id81Fqe-x9Qa6Q,2371
-gllm_inference/schema/__init__.pyi,sha256=-ldt0xJQJirVNdwLFev3bmzmFRw9HSUWBRmmIVH7uyU,1251
+gllm_inference/schema/__init__.pyi,sha256=Mg9aKyvShNaB4XmqLWcZZ0arSNJhT2g1hhIqP1IBuaM,1376
 gllm_inference/schema/attachment.pyi,sha256=9zgAjGXBjLfzPGaKi68FMW6b5mXdEA352nDe-ynOSvY,3385
 gllm_inference/schema/code_exec_result.pyi,sha256=WQ-ARoGM9r6nyRX-A0Ro1XKiqrc9R3jRYXZpu_xo5S4,573
 gllm_inference/schema/enums.pyi,sha256=SQ9mXt8j7uK333uUnUHRs-mkRxf0Z5NCtkAkgQZPIb4,629
@@ -82,13 +82,13 @@ gllm_inference/schema/lm_output.pyi,sha256=WP2LQrY0D03OJtFoaW_dGoJ_-yFUh2HbVlllg
 gllm_inference/schema/message.pyi,sha256=jJV6A0ihEcun2OhzyMtNkiHnf7d6v5R-GdpTBGfJ0AQ,2272
 gllm_inference/schema/model_id.pyi,sha256=3prO19l-FCSecRupe93ruXe91-Xw3GJOpbuQ66bijo0,5368
 gllm_inference/schema/reasoning.pyi,sha256=jbPxkDRHt0Vt-zdcc8lTT1l2hIE1Jm3HIHeNd0hfXGo,577
-gllm_inference/schema/token_usage.pyi,sha256=Eevs8S-yXoM7kQkkzhXHEvORU8DMGzdQynAamqtIoX4,323
+gllm_inference/schema/token_usage.pyi,sha256=WJiGQyz5qatzBK2b-sABLCyTRLCBbAvxCRcqSJOzu-8,3025
 gllm_inference/schema/tool_call.pyi,sha256=OWT9LUqs_xfUcOkPG0aokAAqzLYYDkfnjTa0zOWvugk,403
 gllm_inference/schema/tool_result.pyi,sha256=IJsU3n8y0Q9nFMEiq4RmLEIHueSiim0Oz_DlhKrTqto,287
 gllm_inference/schema/type_alias.pyi,sha256=qAljeBoeQEfT601maGe_mEpXD9inNzbGte1i6joQafc,740
 gllm_inference/utils/__init__.pyi,sha256=RBTWDu1TDPpTd17fixcPYFv2L_vp4-IAOX0IsxgCsD4,299
 gllm_inference/utils/langchain.pyi,sha256=4AwFiVAO0ZpdgmqeC4Pb5NJwBt8vVr0MSUqLeCdTscc,1194
 gllm_inference/utils/validation.pyi,sha256=-RdMmb8afH7F7q4Ao7x6FbwaDfxUHn3hA3WiOgzB-3s,397
-gllm_inference_binary-0.5.9.dist-info/METADATA,sha256=C4ovZJsiSKuRWp_2DbNV57oJwtxC0sI8sPrKXlCHFN4,4531
-gllm_inference_binary-0.5.9.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-gllm_inference_binary-0.5.9.dist-info/RECORD,,
+gllm_inference_binary-0.5.10b3.dist-info/METADATA,sha256=3JPygQCKjjaCWg8h4lzWtx3sLY3KmvD9GUtMnT1dVqQ,4534
+gllm_inference_binary-0.5.10b3.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+gllm_inference_binary-0.5.10b3.dist-info/RECORD,,