# qtimltflite
Source: [https://docs.qualcomm.com/doc/80-70023-50/topic/qtimltflite.html](https://docs.qualcomm.com/doc/80-70023-50/topic/qtimltflite.html)
The qtimltflite plugin shows the LiteRT capabilities (load and run the LiteRT models)
as a GStreamer plugin.
This plugin does the following:
- Accepts a tensor prepared by the preprocessing elements such as
qtimlvconverter.
- Produces the output tensor that's parsed by the postprocessing plugins such as
qtimlvclassification, qtimlvdetection, qtimlvsegmentation, and qtimlvpose.
- Uses the LiteRT library.
- Uses the ION/DMA buffers allocated by the GstMLBufferPool custom buffer pool class
through IOCTL commands to the kernel.
The model and delegate usage is as follows:
- To use a model, provide the absolute path of the model through the model
property.
After it's loaded, the model file provides the input and output
capabilities:
- Number of tensors
- Dimensions
- Type
- If a delegate isn't specified, the plugin executes the model operations on CPU.
- To select a different delegate, set the delegate property accordingly. For more
information, see the [Table : Element properties for qtimltflite](https://docs.qualcomm.com/doc/80-70023-50/topic/qtimltflite.html#qtimltflite__table_txf_2k4_myb).
Figure : GstBuffer workflow with qtimltflite
## Inheritance chain
[GObject](https://docs.gtk.org/gobject/) → [GstObject](https://gstreamer.freedesktop.org/documentation/gstreamer/gstobject.html?gi-language=c) → [GstElement](https://gstreamer.freedesktop.org/documentation/gstreamer/gstelement.html?gi-language=c) → [GstBaseTransform](https://gstreamer.freedesktop.org/documentation/base/gstbasetransform.html?gi-language=c) → GstMLTFLite
The following tables provide information on pad templates and element properties of
qtimltflite. For use cases, see [LiteRT use cases](https://docs.qualcomm.com/doc/80-70023-50/topic/tensorflow-lite-use-cases.html).
## Pad configuration
| Pad Name | Capabilities | Capabilities | Capabilities |
| --- | --- | --- | --- |
| SINK template: 'sink'
| neural-network/tensors | type: | { (string)UINT8, (string)INT32, (string)FLOAT32 } |
## Element configuration
Table : Element properties for qtimltflite
| Property | Description |
| --- | --- |
| name | The name of the object.
flags: readable, writable
String. Default: "mlTFLite0"
|
| parent | The parent of the object.
flags: readable, writable
Object of type "GstObject"
|
| model | Model filename.
flags: readable, writable
String. Default: null
|
| delegate | Delegate part or all graph execution to another executor.
flags: readable, writable
Enum "GstMLTFLiteDelegate" Default: 0, "none"
(0): none - No delegate, CPU is used for all operations
(1): nnapi-dsp - Run the processing on the DSP through the NN API. Unsupported operations fall back on NPU, GPU or CPU
(2): nnapi-gpu - Run the processing on the GPU through the NN API. Unsupported operations fall back on DSP, NPU or CPU
(3): nnapi-npu - Run the processing on the NPU through the NN API. Unsupported operations fall back on DSP, GPU, or CPU
(4): hexagon - Run the processing directly on the Hexagon DSP
(5): gpu - Run the processing directly on the GPU
(6): xnnpack - Run inferences using xnnpack cpu runtime
|
| threads | Number of threads when running operations on the CPU delegate.
flags: readable, writable
Unsigned Integer. Range: 1 - 4 Default: 1
|
**Parent Topic:** [Configure ML plugins](https://docs.qualcomm.com/doc/80-70023-50/topic/inferencing-plugins.html)
Last Published: Mar 27, 2026
[Previous Topic
qtimlsnpe](https://docs.qualcomm.com/bundle/publicresource/80-70023-50/topics/qtimlsnpe.md) [Next Topic
qtimlqnn](https://docs.qualcomm.com/bundle/publicresource/80-70023-50/topics/qtimlqnn.md)