fal.exceptions

from fal.exceptions import (
    FalServerlessException,
    AppException,
    FieldException,
    RequestCancelledException,
    FileTooLargeError,
    AppFileUploadException,
    GPUException,
    GPUOutOfMemoryException,
    CUDAOutOfMemoryException,
    UnauthenticatedException,
    catch_gpu_exceptions,
)

Classes

FalServerlessException

class fal.exceptions.FalServerlessException

Base exception type for fal Serverless related flows and APIs.

Constructor Parameters

Name	Type	Default	Description
`args`	-	-	-
`kwargs`	-	-	-

AppException

class fal.exceptions.AppException

Base exception class for application-specific errors.

Constructor Parameters

Name	Type	Default	Description
`message`	`str`	-	A descriptive message explaining the error.
`status_code`	`int`	-	The HTTP status code associated with the error.

Class Variables

Name	Type	Default	Description
`message`	`str`	-	A descriptive message explaining the error.
`status_code`	`int`	-	The HTTP status code associated with the error.

FieldException

class fal.exceptions.FieldException

Exception raised for errors related to specific fields.

Constructor Parameters

Name	Type	Default	Description
`field`	`str`	-	The field that caused the error.
`message`	`str`	-	A descriptive message explaining the error.
`status_code`	`int`	`422`	The HTTP status code associated with the error. Defaults to 422
`type`	`str`	`'value_error'`	The type of error. Defaults to “value_error”
`billable_units`	`int \| float \| str \| NoneType`	`0`	-

Class Variables

Name	Type	Default	Description
`field`	`str`	-	The field that caused the error.
`message`	`str`	-	A descriptive message explaining the error.
`status_code`	`int`	`422`	The HTTP status code associated with the error. Defaults to 422
`type`	`str`	`'value_error'`	The type of error. Defaults to “value_error”
`billable_units`	`int \| float \| str \| None`	`0`	-

Methods

to_pydantic_format

def to_pydantic_format(self) -> 'dict[str, list[dict]]'

Returns: dict[str, list[dict]]

RequestCancelledException

class fal.exceptions.RequestCancelledException

Exception raised when the request is cancelled by the client.

Constructor Parameters

Name	Type	Default	Description
`message`	`str`	`'Request cancelled by the client.'`	-

Class Variables

Name	Type	Default	Description
`message`	`str`	`'Request cancelled by the client.'`	-

FileTooLargeError

class fal.exceptions.FileTooLargeError

Exception raised when the file is too large.

Constructor Parameters

Name	Type	Default	Description
`message`	`str`	`'File is too large.'`	-

Class Variables

Name	Type	Default	Description
`message`	`str`	`'File is too large.'`	-

AppFileUploadException

class fal.exceptions.AppFileUploadException

Raised when file upload fails

Constructor Parameters

Name	Type	Default	Description
`message`	`str`	-	-
`relative_path`	`str`	-	-

Class Variables

Name	Type	Default	Description
`message`	`str`	-	-
`relative_path`	`str`	-	-

GPUException

class fal.exceptions.GPUException

Base exception for GPU-related errors.

Constructor Parameters

Name	Type	Default	Description
`message`	`str`	`'GPU error'`	-
`status_code`	`int`	`503`	-

Class Variables

Name	Type	Default	Description
`message`	`str`	`'GPU error'`	-
`status_code`	`int`	`503`	-

GPUOutOfMemoryException

class fal.exceptions.GPUOutOfMemoryException

Exception raised when a GPU operation runs out of memory.

Constructor Parameters

Name	Type	Default	Description
`message`	`str`	`'GPU error'`	-
`status_code`	`int`	`503`	-

CUDAOutOfMemoryException

class fal.exceptions.CUDAOutOfMemoryException

Exception raised when a CUDA operation runs out of memory.

Constructor Parameters

Name	Type	Default	Description
`message`	`str`	`'CUDA error: out of memory'`	-
`status_code`	`int`	`503`	-

Class Variables

Name	Type	Default	Description
`message`	`str`	`'CUDA error: out of memory'`	-

UnauthenticatedException

class fal.exceptions.UnauthenticatedException

Base exception type for fal Serverless related flows and APIs.

catch_gpu_exceptions

class fal.exceptions.catch_gpu_exceptions

Catch GPU/CUDA exceptions and convert them to HTTP 503 responses. Works as both a context manager and a decorator. Any caught GPU exception (CUDA OOM, cuDNN errors, NVML failures, etc.) is re-raised as a GPU exception with HTTP status 503. Example:

from fal.exceptions import catch_gpu_exceptions

with catch_gpu_exceptions():
    run_inference()

@catch_gpu_exceptions()
def run_inference():
    ...

Constructor Parameters

Name	Type	Default	Description
`args`	-	-	-
`kwargs`	-	-	-

Client Libraries

CLI Reference

Python SDK

fal.exceptions

Classes

FalServerlessException

AppException

FieldException

to_pydantic_format

RequestCancelledException

FileTooLargeError

AppFileUploadException

GPUException

GPUOutOfMemoryException

CUDAOutOfMemoryException

UnauthenticatedException

catch_gpu_exceptions

​Classes

​FalServerlessException

​AppException

​FieldException

​to_pydantic_format

​RequestCancelledException

​FileTooLargeError

​AppFileUploadException

​GPUException

​GPUOutOfMemoryException

​CUDAOutOfMemoryException

​UnauthenticatedException

​catch_gpu_exceptions

Classes

FalServerlessException

AppException

FieldException

to_pydantic_format

RequestCancelledException

FileTooLargeError

AppFileUploadException

GPUException

GPUOutOfMemoryException

CUDAOutOfMemoryException

UnauthenticatedException

catch_gpu_exceptions