gllm-inference-binary 0.5.23__cp312-cp312-win_amd64.whl → 0.5.25__cp312-cp312-win_amd64.whl

gllm_inference/builder/build_em_invoker.pyi

@@ -86,7 +86,7 @@ def build_em_invoker(model_id: str | ModelId, credentials: str | dict[str, Any]
         # Using Azure OpenAI
         ```python
         em_invoker = build_em_invoker(
-            model_id="azure-openai/https://my-resource.openai.azure.com:my-deployment",
+            model_id="azure-openai/https://my-resource.openai.azure.com/openai/v1:my-deployment",
             credentials="azure-api-key"
         )
         ```

gllm_inference/constants.pyi

@@ -1,11 +1,10 @@
 from _typeshed import Incomplete
 
-DEFAULT_AZURE_OPENAI_API_VERSION: str
+AZURE_OPENAI_URL_SUFFIX: str
 DOCUMENT_MIME_TYPES: Incomplete
 GOOGLE_SCOPES: Incomplete
 GRPC_ENABLE_RETRIES_KEY: str
 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/azure_openai_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 DEFAULT_AZURE_OPENAI_API_VERSION as DEFAULT_AZURE_OPENAI_API_VERSION, INVOKER_PROPAGATED_MAX_RETRIES as INVOKER_PROPAGATED_MAX_RETRIES
+from gllm_inference.constants import AZURE_OPENAI_URL_SUFFIX as AZURE_OPENAI_URL_SUFFIX, INVOKER_PROPAGATED_MAX_RETRIES as INVOKER_PROPAGATED_MAX_RETRIES
 from gllm_inference.em_invoker.openai_em_invoker import OpenAIEMInvoker as OpenAIEMInvoker
 from gllm_inference.em_invoker.schema.openai import Key as Key
 from gllm_inference.schema import ModelId as ModelId, ModelProvider as ModelProvider, TruncationConfig as TruncationConfig
@@ -69,7 +69,7 @@ class AzureOpenAIEMInvoker(OpenAIEMInvoker):
         ```
     '''
     client: Incomplete
-    def __init__(self, azure_endpoint: str, azure_deployment: str, api_key: str | None = None, api_version: str = ..., model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, retry_config: RetryConfig | None = None, truncation_config: TruncationConfig | None = None) -> None:
+    def __init__(self, azure_endpoint: str, azure_deployment: str, api_key: str | None = None, api_version: str | None = None, model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, retry_config: RetryConfig | None = None, truncation_config: TruncationConfig | None = None) -> None:
         """Initializes a new instance of the AzureOpenAIEMInvoker class.
 
         Args:
@@ -77,8 +77,7 @@ class AzureOpenAIEMInvoker(OpenAIEMInvoker):
             azure_deployment (str): The deployment name of the Azure OpenAI service.
             api_key (str | None, optional): The API key for authenticating with Azure OpenAI. Defaults to None, in
                 which case the `AZURE_OPENAI_API_KEY` environment variable will be used.
-            api_version (str, optional): The API version of the Azure OpenAI service. Defaults to
-                `DEFAULT_AZURE_OPENAI_API_VERSION`.
+            api_version (str | None, optional): Deprecated parameter to be removed in v0.6. Defaults to None.
             model_kwargs (dict[str, Any] | None, optional): Additional model parameters. Defaults to None.
             default_hyperparameters (dict[str, Any] | None, optional): Default hyperparameters for invoking the model.
                 Defaults to None.

gllm_inference/em_invoker/bedrock_em_invoker.pyi

@@ -3,6 +3,8 @@ from enum import StrEnum
 from gllm_core.utils.retry import RetryConfig as RetryConfig
 from gllm_inference.em_invoker.em_invoker import BaseEMInvoker as BaseEMInvoker
 from gllm_inference.em_invoker.schema.bedrock import InputType as InputType, Key as Key, OutputType as OutputType
+from gllm_inference.exceptions import BaseInvokerError as BaseInvokerError, convert_http_status_to_base_invoker_error as convert_http_status_to_base_invoker_error
+from gllm_inference.exceptions.provider_error_map import BEDROCK_ERROR_MAPPING as BEDROCK_ERROR_MAPPING
 from gllm_inference.schema import ModelId as ModelId, ModelProvider as ModelProvider, TruncationConfig as TruncationConfig, Vector as Vector
 from typing import Any
 

gllm_inference/em_invoker/em_invoker.pyi

@@ -3,7 +3,7 @@ from _typeshed import Incomplete
 from abc import ABC
 from gllm_core.utils.retry import RetryConfig
 from gllm_inference.constants import DOCUMENT_MIME_TYPES as DOCUMENT_MIME_TYPES, INVOKER_DEFAULT_TIMEOUT as INVOKER_DEFAULT_TIMEOUT
-from gllm_inference.exceptions import parse_error_message as parse_error_message
+from gllm_inference.exceptions import BaseInvokerError as BaseInvokerError, convert_to_base_invoker_error as convert_to_base_invoker_error
 from gllm_inference.schema import Attachment as Attachment, AttachmentType as AttachmentType, EMContent as EMContent, ModelId as ModelId, TruncateSide as TruncateSide, TruncationConfig as TruncationConfig, Vector as Vector
 from typing import Any
 

gllm_inference/em_invoker/google_em_invoker.pyi

@@ -3,6 +3,8 @@ from gllm_core.utils.retry import RetryConfig as RetryConfig
 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.exceptions import BaseInvokerError as BaseInvokerError, convert_http_status_to_base_invoker_error as convert_http_status_to_base_invoker_error
+from gllm_inference.exceptions.provider_error_map import GOOGLE_ERROR_MAPPING as GOOGLE_ERROR_MAPPING
 from gllm_inference.schema import ModelId as ModelId, ModelProvider as ModelProvider, TruncationConfig as TruncationConfig, Vector as Vector
 from typing import Any
 

gllm_inference/em_invoker/langchain_em_invoker.pyi

@@ -3,6 +3,8 @@ from gllm_core.utils.retry import RetryConfig
 from gllm_inference.constants import INVOKER_DEFAULT_TIMEOUT as INVOKER_DEFAULT_TIMEOUT, INVOKER_PROPAGATED_MAX_RETRIES as INVOKER_PROPAGATED_MAX_RETRIES
 from gllm_inference.em_invoker.em_invoker import BaseEMInvoker as BaseEMInvoker
 from gllm_inference.em_invoker.schema.langchain import Key as Key
+from gllm_inference.exceptions import BaseInvokerError as BaseInvokerError, InvokerRuntimeError as InvokerRuntimeError, build_debug_info as build_debug_info
+from gllm_inference.exceptions.provider_error_map import ALL_PROVIDER_ERROR_MAPPINGS as ALL_PROVIDER_ERROR_MAPPINGS, LANGCHAIN_ERROR_CODE_MAPPING as LANGCHAIN_ERROR_CODE_MAPPING
 from gllm_inference.schema import ModelId as ModelId, ModelProvider as ModelProvider, TruncationConfig as TruncationConfig, Vector as Vector
 from gllm_inference.utils import load_langchain_model as load_langchain_model, parse_model_data as parse_model_data
 from langchain_core.embeddings import Embeddings as Embeddings

gllm_inference/em_invoker/schema/bedrock.pyi

@@ -2,9 +2,11 @@ class Key:
     """Defines valid keys in Bedrock."""
     ACCEPT: str
     CONTENT_TYPE: str
+    HTTP_STATUS_CODE: str
     INPUT_TEXT: str
     INPUT_TYPE: str
     MODEL_ID: str
+    RESPONSE_METADATA: str
     TEXTS: str
 
 class InputType:

gllm_inference/em_invoker/schema/google.pyi

@@ -1,7 +1,9 @@
 class Key:
     """Defines valid keys in Google."""
+    API_KEY: str
     CREDENTIALS: str
     HTTP_OPTIONS: str
     LOCATION: str
     PROJECT: str
     TIMEOUT: str
+    VERTEXAI: str

gllm_inference/em_invoker/schema/langchain.pyi

@@ -1,4 +1,5 @@
 class Key:
     """Defines valid keys in LangChain."""
+    ERROR_CODE: str
     MAX_RETRIES: str
     TIMEOUT: str

gllm_inference/exceptions/__init__.pyi

@@ -1,4 +1,4 @@
-from gllm_inference.exceptions.error_parser import ExtendedHTTPStatus as ExtendedHTTPStatus, HTTP_STATUS_TO_EXCEPTION_MAP as HTTP_STATUS_TO_EXCEPTION_MAP, extract_http_status_code as extract_http_status_code, parse_error_message as parse_error_message
-from gllm_inference.exceptions.exceptions import BaseInvokerError as BaseInvokerError, InvokerRuntimeError as InvokerRuntimeError, ModelNotFoundError as ModelNotFoundError, ProviderAuthError as ProviderAuthError, ProviderInternalError as ProviderInternalError, ProviderInvalidArgsError as ProviderInvalidArgsError, ProviderOverloadedError as ProviderOverloadedError, ProviderRateLimitError as ProviderRateLimitError
+from gllm_inference.exceptions.error_parser import _get_exception_key as _get_exception_key, build_debug_info as build_debug_info, convert_http_status_to_base_invoker_error as convert_http_status_to_base_invoker_error, convert_to_base_invoker_error as convert_to_base_invoker_error
+from gllm_inference.exceptions.exceptions import APIConnectionError as APIConnectionError, APITimeoutError as APITimeoutError, BaseInvokerError as BaseInvokerError, InvokerRuntimeError as InvokerRuntimeError, ModelNotFoundError as ModelNotFoundError, ProviderAuthError as ProviderAuthError, ProviderConflictError as ProviderConflictError, ProviderInternalError as ProviderInternalError, ProviderInvalidArgsError as ProviderInvalidArgsError, ProviderOverloadedError as ProviderOverloadedError, ProviderRateLimitError as ProviderRateLimitError
 
-__all__ = ['HTTP_STATUS_TO_EXCEPTION_MAP', 'BaseInvokerError', 'ExtendedHTTPStatus', 'InvokerRuntimeError', 'ModelNotFoundError', 'ProviderAuthError', 'ProviderInternalError', 'ProviderInvalidArgsError', 'ProviderOverloadedError', 'ProviderRateLimitError', 'extract_http_status_code', 'parse_error_message']
+__all__ = ['_get_exception_key', 'APIConnectionError', 'APITimeoutError', 'BaseInvokerError', 'InvokerRuntimeError', 'ModelNotFoundError', 'ProviderAuthError', 'ProviderConflictError', 'ProviderInternalError', 'ProviderInvalidArgsError', 'ProviderOverloadedError', 'ProviderRateLimitError', 'build_debug_info', 'convert_http_status_to_base_invoker_error', 'convert_to_base_invoker_error']

gllm_inference/exceptions/error_parser.pyi

@@ -1,48 +1,41 @@
-from enum import IntEnum
-from gllm_inference.constants import HTTP_STATUS_CODE_PATTERNS as HTTP_STATUS_CODE_PATTERNS
-from gllm_inference.exceptions.exceptions import BaseInvokerError as BaseInvokerError, InvokerRuntimeError as InvokerRuntimeError, ModelNotFoundError as ModelNotFoundError, ProviderAuthError as ProviderAuthError, ProviderInternalError as ProviderInternalError, ProviderInvalidArgsError as ProviderInvalidArgsError, ProviderOverloadedError as ProviderOverloadedError, ProviderRateLimitError as ProviderRateLimitError
-from gllm_inference.schema import ErrorResponse as ErrorResponse
+from gllm_inference.exceptions.exceptions import BaseInvokerError as BaseInvokerError, InvokerRuntimeError as InvokerRuntimeError
+from gllm_inference.exceptions.provider_error_map import ALL_PROVIDER_ERROR_MAPPINGS as ALL_PROVIDER_ERROR_MAPPINGS, HTTP_STATUS_TO_EXCEPTION_MAP as HTTP_STATUS_TO_EXCEPTION_MAP
 from typing import Any
 
-class ExtendedHTTPStatus(IntEnum):
-    """HTTP status codes outside of the standard HTTPStatus enum.
+def build_debug_info(error: Any, class_name: str) -> dict[str, Any]:
+    """Build debug information for an error.
 
-    Attributes:
-        SERVICE_OVERLOADED (int): HTTP status code for service overloaded.
-    """
-    SERVICE_OVERLOADED = 529
-
-HTTP_STATUS_TO_EXCEPTION_MAP: dict[int, type[BaseInvokerError]]
+    Args:
+        error (Any): The error to extract debug information from.
+        class_name (str): The name of the class that raised the error.
 
-def extract_http_status_code(response: ErrorResponse) -> int | None:
-    '''Extract HTTP status code from error message.
+    Returns:
+        dict[str, Any]: A dictionary containing debug information about the error.
+    """
+def convert_http_status_to_base_invoker_error(error: Exception, invoker: BaseEMInvoker | BaseLMInvoker, status_code_extractor: callable = None, provider_error_mapping: dict[str, type[BaseInvokerError]] = ...) -> BaseInvokerError:
+    """Extract provider error with HTTP status code fallback pattern.
 
-    This function extracts the HTTP status code from the error message. For example,
-    if the error message is "Error code: 401 - Invalid API key", "HTTP 429 Rate limit exceeded",
-    or "status: 500 Internal server error", the function will return "401", "429", or "500" respectively.
+    This function implements the common pattern used by Bedrock and Google invokers
+    where they first try to extract HTTP status codes, then fall back to provider-specific
+    error mappings based on exception keys.
 
     Args:
-        response (ErrorResponse): The response object or error message containing HTTP status code.
+        error (Exception): The error to convert.
+        invoker (BaseEMInvoker | BaseLMInvoker): The invoker instance that raised the error.
+        status_code_extractor (callable): Function to extract status code from error.
+        provider_error_mapping (dict): Provider-specific error mapping dictionary.
 
     Returns:
-        int | None: The extracted HTTP status code, or None if not found.
-    '''
-def parse_error_message(class_name: str, error: Any) -> BaseInvokerError:
-    """Parse error from different AI providers and return appropriate exception type.
-
-    This function analyzes the error message and HTTP status code to determine
-    the most appropriate exception type to return.
+        BaseInvokerError: The converted error.
+    """
+def convert_to_base_invoker_error(error: Exception, invoker: BaseEMInvoker | BaseLMInvoker) -> BaseInvokerError:
+    """Convert provider error into BaseInvokerError.
 
     Args:
-        class_name (str): Class name to include in the error message for clarity.
-        error (Any): The error object or message from the AI provider.
-            Can be an Exception object, Response object, ClientResponse object, string, or dict
-            that might contain HTTP status information.
+        error (Exception): The error to convert.
+        invoker (BaseEMInvoker | BaseLMInvoker): The invoker instance that raised the error.
 
     Returns:
-        BaseInvokerError: The appropriate exception instance based on error analysis.
+        BaseInvokerError: The converted error.
 
-    Raises:
-        CancelledError: If the original error is a CancelledError.
-        TimeoutError: If the original error is a TimeoutError.
     """

gllm_inference/exceptions/exceptions.pyi

@@ -22,10 +22,7 @@ class BaseInvokerError(Exception):
         """
 
 class ProviderInvalidArgsError(BaseInvokerError):
-    """Exception for bad or malformed requests, invalid parameters or structure.
-
-    Corresponds to HTTP 400 status code.
-    """
+    """Exception for bad or malformed requests, invalid parameters or structure."""
     def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
         """Initialize ProviderInvalidArgsError.
 
@@ -36,10 +33,7 @@ class ProviderInvalidArgsError(BaseInvokerError):
         """
 
 class ProviderAuthError(BaseInvokerError):
-    """Exception for authorization failures due to API key issues.
-
-    Corresponds to HTTP 401-403 status codes.
-    """
+    """Exception for authorization failures due to API key issues."""
     def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
         """Initialize ProviderAuthError.
 
@@ -50,10 +44,7 @@ class ProviderAuthError(BaseInvokerError):
         """
 
 class ProviderRateLimitError(BaseInvokerError):
-    """Exception for rate limit violations.
-
-    Corresponds to HTTP 429 status code.
-    """
+    """Exception for rate limit violations."""
     def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
         """Initialize ProviderRateLimitError.
 
@@ -64,10 +55,7 @@ class ProviderRateLimitError(BaseInvokerError):
         """
 
 class ProviderInternalError(BaseInvokerError):
-    """Exception for unexpected server-side errors.
-
-    Corresponds to HTTP 500 status code.
-    """
+    """Exception for unexpected server-side errors."""
     def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
         """Initialize ProviderInternalError.
 
@@ -78,10 +66,7 @@ class ProviderInternalError(BaseInvokerError):
         """
 
 class ProviderOverloadedError(BaseInvokerError):
-    """Exception for when the engine is currently overloaded.
-
-    Corresponds to HTTP 503, 529 status codes.
-    """
+    """Exception for when the engine is currently overloaded."""
     def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
         """Initialize ProviderOverloadedError.
 
@@ -92,10 +77,7 @@ class ProviderOverloadedError(BaseInvokerError):
         """
 
 class ModelNotFoundError(BaseInvokerError):
-    """Exception for model not found errors.
-
-    Corresponds to HTTP 404 status code.
-    """
+    """Exception for model not found errors."""
     def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
         """Initialize ModelNotFoundError.
 
@@ -105,11 +87,41 @@ class ModelNotFoundError(BaseInvokerError):
                 Defaults to None.
         """
 
-class InvokerRuntimeError(BaseInvokerError):
-    """Exception for runtime errors that occur during the invocation of the model.
+class APIConnectionError(BaseInvokerError):
+    """Exception for when the client fails to connect to the model provider."""
+    def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
+        """Initialize APIConnectionError.
+
+        Args:
+            class_name (str): The name of the class that raised the error.
+            debug_info (dict[str, Any] | None, optional): Additional debug information for developers.
+                Defaults to None.
+        """
+
+class APITimeoutError(BaseInvokerError):
+    """Exception for when the request to the model provider times out."""
+    def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
+        """Initialize APITimeoutError.
 
-    Corresponds to HTTP status codes other than the ones defined in HTTP_STATUS_TO_EXCEPTION_MAP.
-    """
+        Args:
+            class_name (str): The name of the class that raised the error.
+            debug_info (dict[str, Any] | None, optional): Additional debug information for developers.
+                Defaults to None.
+        """
+
+class ProviderConflictError(BaseInvokerError):
+    """Exception for when the request to the model provider conflicts."""
+    def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
+        """Initialize ProviderConflictError.
+
+        Args:
+            class_name (str): The name of the class that raised the error.
+            debug_info (dict[str, Any] | None, optional): Additional debug information for developers.
+                Defaults to None.
+        """
+
+class InvokerRuntimeError(BaseInvokerError):
+    """Exception for runtime errors that occur during the invocation of the model."""
     def __init__(self, class_name: str, debug_info: dict[str, Any] | None = None) -> None:
         """Initialize the InvokerRuntimeError.
 

gllm_inference/exceptions/provider_error_map.pyi

@@ -0,0 +1,23 @@
+from _typeshed import Incomplete
+from enum import IntEnum
+from gllm_inference.exceptions.exceptions import APIConnectionError as APIConnectionError, APITimeoutError as APITimeoutError, BaseInvokerError as BaseInvokerError, ModelNotFoundError as ModelNotFoundError, ProviderAuthError as ProviderAuthError, ProviderConflictError as ProviderConflictError, ProviderInternalError as ProviderInternalError, ProviderInvalidArgsError as ProviderInvalidArgsError, ProviderOverloadedError as ProviderOverloadedError, ProviderRateLimitError as ProviderRateLimitError
+
+class ExtendedHTTPStatus(IntEnum):
+    """HTTP status codes outside of the standard HTTPStatus enum.
+
+    Attributes:
+        SERVICE_OVERLOADED (int): HTTP status code for service overloaded.
+    """
+    SERVICE_OVERLOADED = 529
+
+HTTP_STATUS_TO_EXCEPTION_MAP: dict[int, type[BaseInvokerError]]
+ANTHROPIC_ERROR_MAPPING: Incomplete
+BEDROCK_ERROR_MAPPING: Incomplete
+GOOGLE_ERROR_MAPPING: Incomplete
+LANGCHAIN_ERROR_CODE_MAPPING: Incomplete
+LITELLM_ERROR_MAPPING: Incomplete
+OPENAI_ERROR_MAPPING: Incomplete
+TWELVELABS_ERROR_MAPPING: Incomplete
+VOYAGE_ERROR_MAPPING: Incomplete
+GRPC_STATUS_CODE_MAPPING: Incomplete
+ALL_PROVIDER_ERROR_MAPPINGS: Incomplete

gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi

@@ -1,15 +1,13 @@
 from _typeshed import Incomplete
 from gllm_core.schema.tool import Tool as Tool
 from gllm_core.utils.retry import RetryConfig as RetryConfig
-from gllm_inference.constants import INVOKER_PROPAGATED_MAX_RETRIES as INVOKER_PROPAGATED_MAX_RETRIES
+from gllm_inference.constants import AZURE_OPENAI_URL_SUFFIX as AZURE_OPENAI_URL_SUFFIX, INVOKER_PROPAGATED_MAX_RETRIES as INVOKER_PROPAGATED_MAX_RETRIES
 from gllm_inference.lm_invoker.openai_lm_invoker import OpenAILMInvoker as OpenAILMInvoker, ReasoningEffort as ReasoningEffort, ReasoningSummary as ReasoningSummary
 from gllm_inference.lm_invoker.schema.openai import Key as Key
 from gllm_inference.schema import ModelId as ModelId, ModelProvider as ModelProvider, ResponseSchema as ResponseSchema
 from langchain_core.tools import Tool as LangChainTool
 from typing import Any
 
-URL_SUFFIX: str
-
 class AzureOpenAILMInvoker(OpenAILMInvoker):
     '''A language model invoker to interact with Azure OpenAI language models.
 

gllm_inference/lm_invoker/bedrock_lm_invoker.pyi

@@ -2,6 +2,8 @@ from _typeshed import Incomplete
 from gllm_core.event import EventEmitter as EventEmitter
 from gllm_core.schema.tool import Tool as Tool
 from gllm_core.utils.retry import RetryConfig as RetryConfig
+from gllm_inference.exceptions import BaseInvokerError as BaseInvokerError, convert_http_status_to_base_invoker_error as convert_http_status_to_base_invoker_error
+from gllm_inference.exceptions.provider_error_map import BEDROCK_ERROR_MAPPING as BEDROCK_ERROR_MAPPING
 from gllm_inference.lm_invoker.lm_invoker import BaseLMInvoker as BaseLMInvoker
 from gllm_inference.lm_invoker.schema.bedrock import InputType as InputType, Key as Key, OutputType as OutputType
 from gllm_inference.schema import Attachment as Attachment, AttachmentType as AttachmentType, LMOutput as LMOutput, Message as Message, ModelId as ModelId, ModelProvider as ModelProvider, ResponseSchema as ResponseSchema, TokenUsage as TokenUsage, ToolCall as ToolCall, ToolResult as ToolResult

gllm_inference/lm_invoker/google_lm_invoker.pyi

@@ -3,6 +3,8 @@ 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, SECONDS_TO_MILLISECONDS as SECONDS_TO_MILLISECONDS
+from gllm_inference.exceptions import BaseInvokerError as BaseInvokerError, convert_http_status_to_base_invoker_error as convert_http_status_to_base_invoker_error
+from gllm_inference.exceptions.provider_error_map import GOOGLE_ERROR_MAPPING as GOOGLE_ERROR_MAPPING
 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

gllm_inference/lm_invoker/langchain_lm_invoker.pyi

@@ -3,6 +3,8 @@ from gllm_core.event import EventEmitter as EventEmitter
 from gllm_core.schema.tool import Tool as Tool
 from gllm_core.utils.retry import RetryConfig
 from gllm_inference.constants import INVOKER_DEFAULT_TIMEOUT as INVOKER_DEFAULT_TIMEOUT, INVOKER_PROPAGATED_MAX_RETRIES as INVOKER_PROPAGATED_MAX_RETRIES
+from gllm_inference.exceptions import BaseInvokerError as BaseInvokerError, InvokerRuntimeError as InvokerRuntimeError, build_debug_info as build_debug_info
+from gllm_inference.exceptions.provider_error_map import ALL_PROVIDER_ERROR_MAPPINGS as ALL_PROVIDER_ERROR_MAPPINGS, LANGCHAIN_ERROR_CODE_MAPPING as LANGCHAIN_ERROR_CODE_MAPPING
 from gllm_inference.lm_invoker.lm_invoker import BaseLMInvoker as BaseLMInvoker
 from gllm_inference.lm_invoker.schema.langchain import InputType as InputType, Key as Key
 from gllm_inference.schema import Attachment as Attachment, AttachmentType as AttachmentType, LMOutput as LMOutput, Message as Message, MessageRole as MessageRole, ModelId as ModelId, ModelProvider as ModelProvider, ResponseSchema as ResponseSchema, TokenUsage as TokenUsage, ToolCall as ToolCall, ToolResult as ToolResult

gllm_inference/lm_invoker/lm_invoker.pyi

@@ -5,7 +5,7 @@ from gllm_core.event import EventEmitter as EventEmitter
 from gllm_core.schema.tool import Tool
 from gllm_core.utils import RetryConfig
 from gllm_inference.constants import DOCUMENT_MIME_TYPES as DOCUMENT_MIME_TYPES, INVOKER_DEFAULT_TIMEOUT as INVOKER_DEFAULT_TIMEOUT
-from gllm_inference.exceptions import parse_error_message as parse_error_message
+from gllm_inference.exceptions import BaseInvokerError as BaseInvokerError, convert_to_base_invoker_error as convert_to_base_invoker_error
 from gllm_inference.schema import Attachment as Attachment, AttachmentType as AttachmentType, EmitDataType as EmitDataType, LMOutput as LMOutput, Message as Message, MessageContent as MessageContent, MessageRole as MessageRole, ModelId as ModelId, Reasoning as Reasoning, ResponseSchema as ResponseSchema, ToolCall as ToolCall, ToolResult as ToolResult
 from langchain_core.tools import Tool as LangChainTool
 from typing import Any

gllm_inference/lm_invoker/schema/bedrock.pyi

@@ -5,8 +5,11 @@ class Key:
     CONTENT_BLOCK_INDEX: str
     DELTA: str
     DESCRIPTION: str
+    ERROR: str
+    CODE: str
     FORMAT: str
     FUNCTION: str
+    HTTP_STATUS_CODE: str
     INFERENCE_CONFIG: str
     INPUT: str
     INPUT_SCHEMA: str
@@ -14,9 +17,11 @@ class Key:
     JSON: str
     MESSAGE: str
     NAME: str
+    RESPONSE: str
     OUTPUT: str
     OUTPUT_TOKENS: str
     PARAMETERS: str
+    RESPONSE_METADATA: str
     ROLE: str
     SOURCE: str
     START: str

gllm_inference/lm_invoker/schema/langchain.pyi

@@ -1,6 +1,7 @@
 class Key:
     """Defines valid keys in LangChain."""
     ARGS: str
+    ERROR_CODE: str
     FINISH_REASON: str
     ID: str
     IMAGE_URL: str

gllm_inference/lm_invoker/xai_lm_invoker.pyi

@@ -3,6 +3,8 @@ from gllm_core.event import EventEmitter as EventEmitter
 from gllm_core.schema.tool import Tool as Tool
 from gllm_core.utils.retry import RetryConfig as RetryConfig
 from gllm_inference.constants import GRPC_ENABLE_RETRIES_KEY as GRPC_ENABLE_RETRIES_KEY, INVOKER_PROPAGATED_MAX_RETRIES as INVOKER_PROPAGATED_MAX_RETRIES
+from gllm_inference.exceptions import BaseInvokerError as BaseInvokerError, InvokerRuntimeError as InvokerRuntimeError, build_debug_info as build_debug_info
+from gllm_inference.exceptions.provider_error_map import GRPC_STATUS_CODE_MAPPING as GRPC_STATUS_CODE_MAPPING
 from gllm_inference.lm_invoker.lm_invoker import BaseLMInvoker as BaseLMInvoker
 from gllm_inference.lm_invoker.schema.xai import Key as Key, ReasoningEffort as ReasoningEffort
 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

gllm_inference/schema/__init__.pyi

@@ -9,6 +9,6 @@ from gllm_inference.schema.reasoning import Reasoning as Reasoning
 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
+from gllm_inference.schema.type_alias import EMContent as EMContent, MessageContent as MessageContent, ResponseSchema as ResponseSchema, Vector as Vector
 
-__all__ = ['Attachment', 'AttachmentType', 'CodeExecResult', 'EMContent', 'EmitDataType', 'ErrorResponse', 'InputTokenDetails', 'MessageContent', 'LMOutput', 'ModelId', 'ModelProvider', 'Message', 'MessageRole', 'OutputTokenDetails', 'Reasoning', 'ResponseSchema', 'TokenUsage', 'ToolCall', 'ToolResult', 'TruncateSide', 'TruncationConfig', 'Vector']
+__all__ = ['Attachment', 'AttachmentType', 'CodeExecResult', 'EMContent', 'EmitDataType', 'InputTokenDetails', 'MessageContent', 'LMOutput', 'ModelId', 'ModelProvider', 'Message', 'MessageRole', 'OutputTokenDetails', 'Reasoning', 'ResponseSchema', 'TokenUsage', 'ToolCall', 'ToolResult', 'TruncateSide', 'TruncationConfig', 'Vector']

gllm_inference/schema/type_alias.pyi

@@ -1,14 +1,10 @@
-from aiohttp import ClientResponse
 from gllm_inference.schema.attachment import Attachment as Attachment
 from gllm_inference.schema.reasoning import Reasoning as Reasoning
 from gllm_inference.schema.tool_call import ToolCall as ToolCall
 from gllm_inference.schema.tool_result import ToolResult as ToolResult
-from httpx import Response as HttpxResponse
 from pydantic import BaseModel
-from requests import Response
 from typing import Any
 
-ErrorResponse = Response | HttpxResponse | ClientResponse | str | dict[str, Any]
 ResponseSchema = dict[str, Any] | type[BaseModel]
 MessageContent = str | Attachment | ToolCall | ToolResult | Reasoning
 EMContent = str | Attachment | tuple[str | Attachment, ...]

gllm_inference.pyi

@@ -46,9 +46,12 @@ import gllm_inference.schema.TruncationConfig
 import openai
 import asyncio
 import enum
+import gllm_inference.exceptions.BaseInvokerError
+import gllm_inference.exceptions.convert_http_status_to_base_invoker_error
 import gllm_inference.schema.Vector
 import aioboto3
-import gllm_inference.exceptions.parse_error_message
+import asyncio.CancelledError
+import gllm_inference.exceptions.convert_to_base_invoker_error
 import gllm_inference.schema.Attachment
 import gllm_inference.schema.AttachmentType
 import gllm_inference.schema.EMContent
@@ -62,6 +65,8 @@ import concurrent.futures
 import concurrent.futures.ThreadPoolExecutor
 import langchain_core
 import langchain_core.embeddings
+import gllm_inference.exceptions.InvokerRuntimeError
+import gllm_inference.exceptions.build_debug_info
 import gllm_inference.utils.load_langchain_model
 import gllm_inference.utils.parse_model_data
 import io
@@ -71,13 +76,8 @@ import base64
 import sys
 import voyageai
 import voyageai.client_async
-import asyncio.CancelledError
-import asyncio.TimeoutError
 import http
 import http.HTTPStatus
-import aiohttp
-import requests
-import gllm_inference.schema.ErrorResponse
 import gllm_core.constants
 import gllm_core.event
 import gllm_core.schema
@@ -95,6 +95,7 @@ import anthropic
 import gllm_inference.schema.MessageRole
 import langchain_core.language_models
 import langchain_core.messages
+import gllm_inference.exceptions._get_exception_key
 import litellm
 import inspect
 import time
@@ -116,6 +117,7 @@ import uuid
 import pathlib
 import filetype
 import magic
+import requests
 import binascii
 import fnmatch
 import importlib

{gllm_inference_binary-0.5.23.dist-info → gllm_inference_binary-0.5.25.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gllm-inference-binary
-Version: 0.5.23
+Version: 0.5.25
 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-0.5.23.dist-info → gllm_inference_binary-0.5.25.dist-info}/RECORD

@@ -1,6 +1,6 @@
 gllm_inference/__init__.pyi,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gllm_inference/builder/__init__.pyi,sha256=-bw1uDx7CAM7pkvjvb1ZXku9zXlQ7aEAyC83KIn3bz8,506
-gllm_inference/builder/build_em_invoker.pyi,sha256=cBqifw0frhYZinDndeCjqDVqv7oeW728-i5nH4JMyhk,6010
+gllm_inference/builder/build_em_invoker.pyi,sha256=Mh1vRoJhsqc8hX4jUdopV14Fn44ql27NB7xbGjoHJtE,6020
 gllm_inference/builder/build_lm_invoker.pyi,sha256=p63iuVBOOpNizItGK6HDxYDrgXdovtfSe0VrvrEd-PA,7047
 gllm_inference/builder/build_lm_request_processor.pyi,sha256=0pJINCP4nnXVwuhIbhsaiwzjX8gohQt2oqXFZhTFSUs,4584
 gllm_inference/builder/build_output_parser.pyi,sha256=sgSTrzUmSRxPzUUum0fDU7A3NXYoYhpi6bEx4Q2XMnA,965
@@ -8,51 +8,52 @@ gllm_inference/catalog/__init__.pyi,sha256=HWgPKWIzprpMHRKe_qN9BZSIQhVhrqiyjLjIX
 gllm_inference/catalog/catalog.pyi,sha256=eWPqgQKi-SJGHabi_XOTEKpAj96OSRypKsb5ZEC1VWU,4911
 gllm_inference/catalog/lm_request_processor_catalog.pyi,sha256=GemCEjFRHNChtNOfbyXSVsJiA3klOCAe_X11fnymhYs,5540
 gllm_inference/catalog/prompt_builder_catalog.pyi,sha256=iViWB4SaezzjQY4UY1YxeoXUNxqxa2cTJGaD9JSx4Q8,3279
-gllm_inference/constants.pyi,sha256=kvYdaD0afopdfvijkyTSq1e5dsUKiN232wA9KlcPGs8,325
+gllm_inference/constants.pyi,sha256=KQmondDEkHK2P249ymmce3SdutVrx8kYm4v1eTCkW9U,277
 gllm_inference/em_invoker/__init__.pyi,sha256=pmbsjmsqXwfe4WPykMnrmasKrYuylJWnf2s0pbo0ioM,997
-gllm_inference/em_invoker/azure_openai_em_invoker.pyi,sha256=g1I3Aexg5VeDeU_zbZWCVgca2fhrUztVrpbzS5GBBYI,5072
-gllm_inference/em_invoker/bedrock_em_invoker.pyi,sha256=EbXyj_U0NK9QSnq9HSeCHJ1Hw7xg2Twqj4wcbkHuvng,5560
-gllm_inference/em_invoker/em_invoker.pyi,sha256=l_jnFRrfoVatVwKawpPA018bM0U6wMc8j_DVxkL8T4s,5133
-gllm_inference/em_invoker/google_em_invoker.pyi,sha256=DH_ddq07EfUgv5L0OTZVOhg-p3CqEpcWAjmCYJsSljM,6684
+gllm_inference/em_invoker/azure_openai_em_invoker.pyi,sha256=SfJPC_PJGiEfWS9JH5kRQPJztsR7jRhwVuETqdY-JsQ,5021
+gllm_inference/em_invoker/bedrock_em_invoker.pyi,sha256=UqodtpDmE7fEgpctXEETIlZGorX9i1lmmuTvGaJke6o,5829
+gllm_inference/em_invoker/em_invoker.pyi,sha256=YDYJ8TGScsz5Gg-OBnEENN1tI1RYvwoddypxUr6SAWw,5191
+gllm_inference/em_invoker/google_em_invoker.pyi,sha256=q69kdVuE44ZqziQ8BajFYZ1tYn-MPjKjzXS9cRh4oAo,6951
 gllm_inference/em_invoker/langchain/__init__.pyi,sha256=aOTlRvS9aG1tBErjsmhe75s4Sq-g2z9ArfGqNW7QyEs,151
 gllm_inference/em_invoker/langchain/em_invoker_embeddings.pyi,sha256=BBSDazMOckO9Aw17tC3LGUTPqLb01my1xUZLtKZlwJY,3388
-gllm_inference/em_invoker/langchain_em_invoker.pyi,sha256=vQO5yheucM5eb7xWcwb4U7eGXASapwgOFC_SZdyysHA,3207
+gllm_inference/em_invoker/langchain_em_invoker.pyi,sha256=nhX6LynrjhfySEt_44OlLoSBd15hoz3giWyNM9CYLKY,3544
 gllm_inference/em_invoker/openai_compatible_em_invoker.pyi,sha256=zEYOBDXKQhvcMGer9DYDu50_3KRDjYyN8-JgpBIFPOI,5456
 gllm_inference/em_invoker/openai_em_invoker.pyi,sha256=0TDIQa-5UwsPcVxgkze-QJJWrt-ToakAKbuAk9TW5SM,4746
 gllm_inference/em_invoker/schema/__init__.pyi,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-gllm_inference/em_invoker/schema/bedrock.pyi,sha256=6xP5T5jxane_Ecrb-6zf_X678Tj3svQUNy0SgHQArRM,443
-gllm_inference/em_invoker/schema/google.pyi,sha256=lPzJ-f18qVar6dctdN4eQWrxWrOFHC9zJ4cuLXXMytw,153
-gllm_inference/em_invoker/schema/langchain.pyi,sha256=JPEqA6naKL64jpW8NEKsEP-V9STY2h8wvyDsFtFEHos,96
+gllm_inference/em_invoker/schema/bedrock.pyi,sha256=HoNgVi0T21aFd1JrCnSLu4yryv8k8RnYdR3-tIdHFgA,498
+gllm_inference/em_invoker/schema/google.pyi,sha256=bzdtu4DFH2kATLybIeNl_Lznj99H-6u2Fvx3Zx52oZg,190
+gllm_inference/em_invoker/schema/langchain.pyi,sha256=SZ13HDcvAOGmDTi2b72H6Y1J5GePR21JdnM6gYrwcGs,117
 gllm_inference/em_invoker/schema/openai.pyi,sha256=rNRqN62y5wHOKlr4T0n0m41ikAnSrD72CTnoHxo6kEM,146
 gllm_inference/em_invoker/schema/openai_compatible.pyi,sha256=A9MOeBhI-IPuvewOk4YYOAGtgyKohERx6-9cEYtbwvs,157
 gllm_inference/em_invoker/schema/twelvelabs.pyi,sha256=D3F9_1F-UTzE6Ymxj6u0IFdL6OFVGlc7noZJr3iuA6I,389
 gllm_inference/em_invoker/schema/voyage.pyi,sha256=Aqvu6mhFkNb01aXAI5mChLKIgEnFnr-jNKq1lVWB54M,304
 gllm_inference/em_invoker/twelevelabs_em_invoker.pyi,sha256=MMVgSnjMXksdhSDXIi3vOULIXnjbhtq19eR5LPnUmGo,5446
 gllm_inference/em_invoker/voyage_em_invoker.pyi,sha256=vdB_qS8QKrCcb-HtXwKZS4WW1R1wGzpMBFmOKC39sjU,5619
-gllm_inference/exceptions/__init__.pyi,sha256=2F05RytXZIKaOJScb1pD0O0bATIQHVeEAYYNX4y5N2A,981
-gllm_inference/exceptions/error_parser.pyi,sha256=ggmh8DJXdwFJInNLrP24WVJt_4raxbAVxzXRQgBpndA,2441
-gllm_inference/exceptions/exceptions.pyi,sha256=ViXvIzm7tLcstjqfwC6nPziDg0UAmoUAWZVWrAJyp3w,4763
+gllm_inference/exceptions/__init__.pyi,sha256=nXOqwsuwUgsnBcJEANVuxbZ1nDfcJ6-pKUfKeZwltkk,1218
+gllm_inference/exceptions/error_parser.pyi,sha256=4aiJZhBzBOqlhdmpvaCvildGy7_XxlJzQpe3PzGt8eE,2040
+gllm_inference/exceptions/exceptions.pyi,sha256=6y3ECgHAStqMGgQv8Dv-Ui-5PDD07mSj6qaRZeSWea4,5857
+gllm_inference/exceptions/provider_error_map.pyi,sha256=4AsAgbXAh91mxEW2YiomEuhBoeSNeAIo9WbT9WK8gQk,1233
 gllm_inference/lm_invoker/__init__.pyi,sha256=eE_HDCl9A135mi6mtIV55q-T9J1O8OpbMcqWuny3w9A,1214
 gllm_inference/lm_invoker/anthropic_lm_invoker.pyi,sha256=85uvShLv4-eiGOpTMgwWpQGZXPW6XaB6GrexBmxg_sQ,15200
-gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi,sha256=Ek7pZYaSWWFdPKI6iPKiICvZxN7xCVglQleTSSwW5ok,14799
-gllm_inference/lm_invoker/bedrock_lm_invoker.pyi,sha256=ae5P_9sjtcOgMIUaRchvp8F0FujoeP4e2F_OoHSe_go,12655
+gllm_inference/lm_invoker/azure_openai_lm_invoker.pyi,sha256=Wzw6We1KwLcWW9-4tGGMZoPnnHSKofOmCuqoddTHr2Q,14832
+gllm_inference/lm_invoker/bedrock_lm_invoker.pyi,sha256=HzpDRIhe4-XRj9n43bmsXQHxPwx5lcnetxIe5EMbHIE,12924
 gllm_inference/lm_invoker/datasaur_lm_invoker.pyi,sha256=c4H3TOz0LIhWjokCCdQ4asiwQR4_LPyaimo4RAqU9es,9369
-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/google_lm_invoker.pyi,sha256=IPmVAFTtZGvBDb-veoeCq8u7R9chKU958vJoBlWbIvE,17369
+gllm_inference/lm_invoker/langchain_lm_invoker.pyi,sha256=kH28ELOda6_5rNRDHSNZOicEd90jCPQnf2pLewZdW5s,13859
 gllm_inference/lm_invoker/litellm_lm_invoker.pyi,sha256=HHwW7i8ryXHI23JZQwscyva6aPmPOB13Muhf7gaaMUM,13376
-gllm_inference/lm_invoker/lm_invoker.pyi,sha256=Sd-ywxgPcIzyI5eA7XoqdkYG9hntEnihJfj6Ack7qr0,7975
+gllm_inference/lm_invoker/lm_invoker.pyi,sha256=B00siZZ7F3i2GuU4nQk3xA8d-h_b37ADzyYBoXarbPA,8033
 gllm_inference/lm_invoker/openai_compatible_lm_invoker.pyi,sha256=JemahodhaUsC2gsI7YSxnW4X3uX1cU4YCFdIvdWWY88,15203
 gllm_inference/lm_invoker/openai_lm_invoker.pyi,sha256=VFMvYXuwMuUHarsu5Xz7tKF6Bx6Ket5HaXZ4-7AtBY0,20011
 gllm_inference/lm_invoker/schema/__init__.pyi,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gllm_inference/lm_invoker/schema/anthropic.pyi,sha256=lGJ7xYLchdtv6003Is4GcaKiGdbmIOAzQsaldKG0Aww,1041
-gllm_inference/lm_invoker/schema/bedrock.pyi,sha256=H3attoGWhBA725W4FpXw7Mty46N9jHKjw9PT-0lMEJs,975
+gllm_inference/lm_invoker/schema/bedrock.pyi,sha256=rB1AWfER2BBKZ5I219211YE2EUFPF25bhzysqjdPgiY,1080
 gllm_inference/lm_invoker/schema/datasaur.pyi,sha256=GLv6XAwKtWyRrX6EsbEufYjkPffHNiEpXwJOn9HqxMA,242
 gllm_inference/lm_invoker/schema/google.pyi,sha256=elXHrUMS46pbTsulk7hBXVVFcT022iD-_U_I590xeV8,529
-gllm_inference/lm_invoker/schema/langchain.pyi,sha256=uEG0DSD0z4L_rDMkBm-TtUy5oTyEHEEJWiLsYvFf1sw,431
+gllm_inference/lm_invoker/schema/langchain.pyi,sha256=2OJOUQPlGdlUbIOTDOyiWDBOMm3MoVX-kU2nK0zQsF0,452
 gllm_inference/lm_invoker/schema/openai.pyi,sha256=2KZkitU0jxFaR6x2AGe1FtawvxtUgTLDffY9T0Iq9yg,2017
 gllm_inference/lm_invoker/schema/openai_compatible.pyi,sha256=fVLRIrOvLJjhY7qPUgC3HRFoOFa7XimWLjr2EOo5qmQ,1226
 gllm_inference/lm_invoker/schema/xai.pyi,sha256=jpC6ZSBDUltzm9GjD6zvSFIPwqizn_ywLnjvwSa7KuU,663
-gllm_inference/lm_invoker/xai_lm_invoker.pyi,sha256=61Jihta4Mm1SZ72aGmXNPFW2g2_TJx7BxjfY_jugvVY,15723
+gllm_inference/lm_invoker/xai_lm_invoker.pyi,sha256=6TwO3KU1DBWoe4UAsz97MY1yKBf-N38WjbrBqCmWCNU,15992
 gllm_inference/model/__init__.pyi,sha256=JKQB0wVSVYD-_tdRkG7N_oEVAKGCcoBw0BUOUMLieFo,602
 gllm_inference/model/em/__init__.pyi,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gllm_inference/model/em/google_em.pyi,sha256=c53H-KNdNOK9ppPLyOSkmCA890eF5FsMd05upkPIzF0,487
@@ -78,7 +79,7 @@ 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=0fy1HyILCVDw6y46E-7tLnQTRYx4ppeRMe0QP6t9Jyw,5990
 gllm_inference/request_processor/uses_lm_mixin.pyi,sha256=LYHq-zLoXEMel1LfVdYv7W3BZ8WtBLo_WWFjRf10Yto,6512
-gllm_inference/schema/__init__.pyi,sha256=bYdXkfqkNAKEr48xaOKKQTbt2zLcCPiLCdSl2UTEIfE,1521
+gllm_inference/schema/__init__.pyi,sha256=7PMhg0y3kKYNSrl5LvHZeHKk6WqDJz4p_rYQ6u3_5cY,1472
 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/config.pyi,sha256=NVmjQK6HipIE0dKSfx12hgIC0O-S1HEcAc-TWlXAF5A,689
@@ -90,14 +91,14 @@ gllm_inference/schema/reasoning.pyi,sha256=jbPxkDRHt0Vt-zdcc8lTT1l2hIE1Jm3HIHeNd
 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=TQiskxG9yUlvj7xEj8X84pNxlhMvhbehujbUPPSVKD0,734
+gllm_inference/schema/type_alias.pyi,sha256=L-V0WxsFQznzAfby3DH8XMHUgjjZxQEsLw8SbhdlXts,540
 gllm_inference/utils/__init__.pyi,sha256=H27RiiFjD6WQHRrYb1-sBnb2aqjVENw5_8-DdAe1k9A,396
 gllm_inference/utils/io_utils.pyi,sha256=Eg7dvHWdXslTKdjh1j3dG50i7r35XG2zTmJ9XXvz4cI,1120
 gllm_inference/utils/langchain.pyi,sha256=4AwFiVAO0ZpdgmqeC4Pb5NJwBt8vVr0MSUqLeCdTscc,1194
 gllm_inference/utils/validation.pyi,sha256=-RdMmb8afH7F7q4Ao7x6FbwaDfxUHn3hA3WiOgzB-3s,397
 gllm_inference.build/.gitignore,sha256=aEiIwOuxfzdCmLZe4oB1JsBmCUxwG8x-u-HBCV9JT8E,1
-gllm_inference.cp312-win_amd64.pyd,sha256=pFLMLpBS5yOsZxT3fwEAOSZ-t8kcRnLVByTDXYf-kqM,2994176
-gllm_inference.pyi,sha256=lTVixRzlC12Joi4kW_vxnux0rLHAUB_3j7RMFOwLK-M,3616
-gllm_inference_binary-0.5.23.dist-info/METADATA,sha256=zuk4QzAoRrrwKmjbfX4jgFotZrE6Olk-behj1vcrRGk,4615
-gllm_inference_binary-0.5.23.dist-info/WHEEL,sha256=4N0hGcnWMI_Ty6ATf4qJqqSl-UNI-Ln828iTWGIywmU,98
-gllm_inference_binary-0.5.23.dist-info/RECORD,,
+gllm_inference.cp312-win_amd64.pyd,sha256=chmSP2dPOzlrqoDWRSzR89vi75CzNscirmC4I2nTMAw,3068928
+gllm_inference.pyi,sha256=Pq6P04np3S3x7juGVCzC5sL2im4MsyligEvahVQNWzM,3820
+gllm_inference_binary-0.5.25.dist-info/METADATA,sha256=EegMUA9X6Hb5cn9UveDy_bEBJCV5oB860sE0GEv65mc,4615
+gllm_inference_binary-0.5.25.dist-info/WHEEL,sha256=4N0hGcnWMI_Ty6ATf4qJqqSl-UNI-Ln828iTWGIywmU,98
+gllm_inference_binary-0.5.25.dist-info/RECORD,,