# qairt.api.executor

## qairt.api.executor.execution\_config

- *class* qairt.api.executor.execution\_config.ExecutionConfig(*\*args: Any*, *\*\*kwargs: Any*)

    - Bases: `AISWBaseModel`

Pydantic class of parameters for model execution

- backend*: Optional[Union[[BackendType](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-configs.html#qairt.api.configs.common.BackendType), str]]*  *= None*

    - Backend to be used for execution.

- batch\_multiplier*: Optional[str]*  *= None*

    - Specifies the value with which the batch value in input and
output tensors dimensions will be multiplied. The modified input and output tensors will be
used only during the execute graphs.

- context\_custom\_configs*: Optional[Union[List[[HtpContextConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-common-backends-htp.html#qairt.api.common.backends.htp.config.HtpContextConfig)], List[[HtpMcpContextConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-common-backends-htp_mcp.html#qairt.api.common.backends.htp_mcp.config.HtpMcpContextConfig)]]]*  *= None*

    - Context configuration options specific to a backend.
Only HTP and HTP MCP backend options are supported.
See <cite>qairt.api.common.backends.htp.config.HtpContextConfig</cite> and
<cite>qairt.api.common.backends.htp_mcp.config.HtpMcpContextConfig</cite> for options

- context\_execute\_custom\_config*: Optional[[ContextExecuteConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-configs.html#qairt.api.configs.common.ContextExecuteConfig)]*  *= None*

    - Context configuration options for setting the priority, peak memory of a context.
See <cite>qairt.api.configs.common.ContextExecuteConfig</cite> for options

- debug*: Optional[bool]*  *= None*

    - Specifies that output from all layers of the network will be saved.

- device\_custom\_configs*: Optional[Union[List[[HtpDeviceConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-common-backends-htp.html#qairt.api.common.backends.htp.config.HtpDeviceConfig)], List[[HtpMcpDeviceConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-common-backends-htp_mcp.html#qairt.api.common.backends.htp_mcp.config.HtpMcpDeviceConfig)]]]*  *= None*

    - Device configuration options specific to a backend.
Only HTP and HTP MCP backend options are supported.
See <cite>qairt.api.common.backends.htp.config.HtpDeviceConfig</cite>
and <cite>qairt.api.common.backends.htp.config.HtpMcpDeviceConfig</cite> for options.

- duration*: Optional[float]*  *= None*

    - Specifies the duration of the graph execution in seconds.
Loops over the input\_list until this amount of time has transpired.

- graph\_custom\_configs*: Optional[List[[HtpGraphConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-common-backends-htp.html#qairt.api.common.backends.htp.config.HtpGraphConfig)]]*  *= None*

    - Graph configuration options specific to a backend.
Currently only used for HTP backend. See [`qairt.api.common.backends.htp.config.HtpGraphConfig`](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-common-backends-htp.html#qairt.api.common.backends.htp.config.HtpGraphConfig)
for options.

- keep\_num\_outputs*: Optional[int]*  *= None*

    - Specifies the number of outputs to be saved. Once the number of outputs reach the limit,
subsequent outputs would be discarded.

- log\_level*: Optional[str]*  *= None*

    - Log level for the executor. Standard logging levels are supported.

- memory\_custom\_config*: Optional[[HtpMemoryConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-common-backends-htp.html#qairt.api.common.backends.htp.config.HtpMemoryConfig)]*  *= None*

    - Memory backend configuration for the compiler. Only HTP backend configurations are supported.

- model\_post\_init(*\_ExecutionConfig\_\_context*)

    - 

- native\_input\_tensor\_names*: Optional[List[str]]*  *= None*

    - List of input tensor names,for which the input files
would be read/parsed in native format. Note that
options use\_native\_input\_data and native\_input\_tensor\_names are mutually exclusive.

- num\_inferences*: Optional[int]*  *= None*

    - Specifies the number of inferences. Loops over the input\_list until
the specified number of inferences has transpired.

- op\_packages*: Optional[List[qti.aisw.tools.core.modules.api.OpPackageIdentifier]]*  *= None*

    - Provide a comma-separated list of op packages, interface
providers, and, optionally, targets to register. Valid values
for target are CPU and HTP.

- optional\_output\_tensor\_names*: Optional[Union[List[str], Dict[str, Optional[List[str]]]]]*  *= None*

    - Controls which optional output tensors to return. Mandatory outputs are always included.
For single-graph models: None returns all optional outputs; an empty list returns no optional outputs,
and a list of names returns only the specified optional outputs.
For multi-graph models: Provide a dict mapping graph names to a output tensor (None, empty list,
or list of names).

- perf\_profile*: Optional[[PerfProfile](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-configs.html#qairt.api.configs.common.PerfProfile)]*  *= None*

    - Specifies performance profile to be used. Valid settings are
low\_balanced, balanced, default, high\_performance, sustained\_high\_performance, burst,
low\_power\_saver, power\_saver, high\_power\_saver, extreme\_power\_saver and system\_settings.

- platform\_options*: Optional[Union[str, Dict[str, str]]]*  *= None*

    - Specifies values to pass as platform options.

- profiling\_level*: Optional[qti.aisw.tools.core.modules.api.ProfilingLevel]*  *= None*

    - options are “basic”, “backend”, “detailed” and “client”.
This field should be set within a profiler context.

- Type

    - Profiling levels

- profiling\_option*: Optional[qti.aisw.tools.core.modules.api.ProfilingOption]*  *= None*

    - “optrace. This field should be set within a profiler context.

- Type

    - Profiling options

- runtime\_custom\_config*: Optional[[AicRuntimeConfig](https://docs.qualcomm.com/doc/80-87189-2/topic/qairt-api-common-backends-aic.html#qairt.api.common.backends.aic.config.AicRuntimeConfig)]*  *= None*

    - Set this field to enable configurations that are passed by the backend to the executor.
Note this option is currently only applicable to the AIC Backend.
Use qairt.api.common.backends.aic.AicRuntimeConfig.list\_config\_options to see valid fields.

- set\_output\_tensors*: Optional[List[str]]*  *= None*

    - List of intermediate output tensor names,
for which the outputs will be written in addition to final graph output tensors.
Note that options debug and set\_output\_tensors are mutually exclusive.

- synchronous*: Optional[bool]*  *= None*

    - Specifies the way graphs should be executed.

- use\_mmap*: Optional[bool]*  *= False*

    - Specifies whether to use mmap for memory allocation.

- use\_native\_input\_data*: Optional[bool]*  *= None*

    - Specifies that the input files will be parsed in the data type
native to the graph. If not specified, input files will be parsed in floating point.
Note that options use\_native\_input\_data and native\_input\_tensor\_names are mutually exclusive.

- use\_native\_output\_data*: Optional[bool]*  *= None*

    - Specifies that the output files will be generated
in the data type native to the graph or in floating point.

Last Published: May 26, 2026

[Previous Topic
qairt.api.converter](https://docs.qualcomm.com/bundle/publicresource/80-87189-2/topics/qairt-api-converter.md) [Next Topic
qairt.api.profiler](https://docs.qualcomm.com/bundle/publicresource/80-87189-2/topics/qairt-api-profiler.md)