Model¶
Represents an AI model loaded from an MXQ file.
This class loads an AI model from an MXQ file and provides functions to launch it on the NPU and perform inference.
NHWC float-to-float inference¶
Performs inference with input and output elements of type float in NHWC (batch N, height H, width W, channels C) or HWC format.
Two input-output type pairs are supported:
std::vector<NDArray<float>>for both input and output- Recommended approach, as
NDArrayallows the qbruntime to avoid unnecessary data copies internally.
- Recommended approach, as
std::vector<float*>for input andstd::vector<std::vector<float>>for output- Provided for user convenience, but results in unavoidable extra copies within the qbruntime.
| Member | Description |
|---|---|
infer ×14 |
Performs inference. |
infer¶
Performs inference.
Parameters
input[in] — A vector ofNDArray<float>. Each NDArray must be in NHWC or HWC format.output[out] — A reference to a vector ofNDArray<float>that will store the inference results.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
This overload differs from the above function in that it directly returns the inference results instead of modifying an output parameter.
Parameters
input[in] — A vector ofNDArray<float>. Each NDArray must be in NHWC or HWC format.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of NDArray<float> containing the inference results.
This overload is provided for convenience but may result in additional data copies within the qbruntime.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in HWC format.output[out] — A reference to a vector of float vectors that will store the inference results.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
This overload is provided for convenience but may result in additional data copies within the qbruntime.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in HWC format.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of float vectors containing the inference results.
StatusCode infer(const std::vector<float *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape)
This overload is provided for convenience but may result in additional data copies within the qbruntime.
Unlike other overloads, this version allows explicitly specifying the shape of each input data, which can be in NHWC or HWC format.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NHWC or HWC format.output[out] — A reference to a vector of float vectors that will store the inference results.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
std::vector<std::vector<float>> infer(const std::vector<float *> &input, const std::vector<std::vector<int64_t>> &shape, StatusCode &sc)
This overload is provided for convenience but may result in additional data copies within the qbruntime.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NHWC or HWC format.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of float vectors containing the inference results.
StatusCode infer(const std::vector<NDArray<float>> &input, std::vector<NDArray<float>> &output, uint32_t cache_size)
This overload supports inference with KV cache.
Note
This function is relevant for LLM models that use KV cache.
Parameters
input[in] — A vector of NDArrays, where each NDArray represents input data in NHWC or HWC format.output[out] — A reference to a vector of NDArrays that will store the inference results.cache_size[in] — The number of tokens accumulated in the KV cache so far.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
StatusCode infer(const std::vector<NDArray<float>> &input, std::vector<NDArray<float>> &output, const std::vector<BatchParam> ¶ms)
This overload is supports inference with BatchParam for BatchLLM.
Parameters
input[in] — A vector of NDArrays, where each NDArray represents input data in NHWC or HWC format.output[out] — A reference to a vector of NDArrays that will store the inference results.params[in] — A vector ofBatchParam, specifying each Batch's information for BatchLLM inference.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
std::vector<NDArray<float>> infer(const std::vector<NDArray<float>> &input, uint32_t cache_size, StatusCode &sc)
This overload supports inference with KV cache.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Note
This function is relevant for LLM models that use KV cache.
Parameters
input[in] — A vector of NDArrays, where each NDArray represents input data in NHWC or HWC format.cache_size[in] — The number of tokens accumulated in the KV cache so far.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of NDArrays containing the inference results.
std::vector<NDArray<float>> infer(const std::vector<NDArray<float>> &input, const std::vector<BatchParam> ¶ms, StatusCode &sc)
This overload is supports inference with BatchParam for BatchLLM.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Parameters
input[in] — A vector of NDArrays, where each NDArray represents input data in NHWC or HWC format.params[in] — A vector ofBatchParam, specifying each Batch's information for BatchLLM inference.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of NDArrays containing the inference results.
StatusCode infer(const std::vector<float *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size)
This overload supports inference with KV cache.
Note
This function is relevant for LLM models that use KV cache.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NHWC or HWC format.output[out] — A reference to a vector of float vectors that will store the inference results.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.cache_size[in] — The number of tokens accumulated in the KV cache so far.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
std::vector<std::vector<float>> infer(const std::vector<float *> &input, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size, StatusCode &sc)
This overload supports inference with KV cache.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Note
This function is relevant for LLM models that use KV cache.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NHWC or HWC format.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.cache_size[in] — The number of tokens accumulated in the KV cache so far.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of float vectors containing the inference results.
StatusCode infer(const std::vector<float *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape, const std::vector<BatchParam> ¶ms)
This overload is supports inference with BatchParam for BatchLLM.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NHWC or HWC format.output[out] — A reference to a vector of float vectors that will store the inference results.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.params[in] — A vector ofBatchParam, specifying each Batch's information for BatchLLM inference.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
std::vector<std::vector<float>> infer(const std::vector<float *> &input, const std::vector<std::vector<int64_t>> &shape, const std::vector<BatchParam> ¶ms, StatusCode &sc)
This overload is supports inference with BatchParam for BatchLLM.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NHWC or HWC format.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.params[in] — A vector ofBatchParam, specifying each Batch's information for BatchLLM inference.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of float vectors containing the inference results.
NCHW float-to-float inference¶
Performs inference with input and output elements of type float in NCHW (batch N, channels C, height H, width W) or CHW format.
Two input-output type pairs are supported:
std::vector<NDArray<float>>for both input and output- Recommended approach, as
NDArrayallows the qbruntime to avoid unnecessary data copies internally.
- Recommended approach, as
std::vector<float*>for input andstd::vector<std::vector<float>>for output- Provided for user convenience, but results in unavoidable extra copies within the qbruntime.
Note
CHW is not the recommended format, as the NPU natively operates on HWC-ordered data. When input is provided in CHW format, it will be transposed internally, introducing additional overhead.
Note
If your data is in HWC format, use Model::infer instead of Model::inferCHW, as it avoids unnecessary format conversion.
| Member | Description |
|---|---|
inferCHW ×10 |
Performs inference. |
inferCHW¶
Performs inference.
Parameters
input[in] — A vector ofNDArray<float>. Each NDArray must be in NCHW or CHW format.output[out] — A reference to a vector ofNDArray<float>that will store the inference results.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
This overload differs from the above function in that it directly returns the inference results instead of modifying an output parameter.
Parameters
input[in] — A vector ofNDArray<float>. Each NDArray must be in NCHW or CHW format.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of NDArray<float> containing the inference results.
This overload is provided for convenience but may result in additional data copies within the qbruntime.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in CHW format.output[out] — A reference to a vector of float vectors that will store the inference results.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
This overload is provided for convenience but may result in additional data copies within the qbruntime.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in CHW format.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of float vectors containing the inference results.
StatusCode inferCHW(const std::vector<float *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape)
This overload is provided for convenience but may result in additional data copies within the qbruntime.
Unlike other overloads, this version allows explicitly specifying the shape of each input data, which can be in NCHW or CHW format.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NCHW or CHW format.output[out] — A reference to a vector of float vectors that will store the inference results.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
std::vector<std::vector<float>> inferCHW(const std::vector<float *> &input, const std::vector<std::vector<int64_t>> &shape, StatusCode &sc)
This overload is provided for convenience but may result in additional data copies within the qbruntime.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NCHW or CHW format.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of float vectors containing the inference results.
StatusCode inferCHW(const std::vector<NDArray<float>> &input, std::vector<NDArray<float>> &output, uint32_t cache_size)
This overload supports inference with KV cache.
Note
This function is relevant for LLM models that use KV cache.
Parameters
input[in] — A vector of NDArrays, where each NDArray represents input data in NCHW or CHW format.output[out] — A reference to a vector of NDArrays that will store the inference results.cache_size[in] — The number of tokens accumulated in the KV cache so far.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
std::vector<NDArray<float>> inferCHW(const std::vector<NDArray<float>> &input, uint32_t cache_size, StatusCode &sc)
This overload supports inference with KV cache.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Note
This function is relevant for LLM models that use KV cache.
Parameters
input[in] — A vector of NDArrays, where each NDArray represents input data in NCHW or CHW format.cache_size[in] — The number of tokens accumulated in the KV cache so far.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of NDArrays containing the inference results.
StatusCode inferCHW(const std::vector<float *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size)
This overload supports inference with KV cache.
Note
This function is relevant for LLM models that use KV cache.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NCHW or CHW format.output[out] — A reference to a vector of float vectors that will store the inference results.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.cache_size[in] — The number of tokens accumulated in the KV cache so far.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
std::vector<std::vector<float>> inferCHW(const std::vector<float *> &input, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size, StatusCode &sc)
This overload supports inference with KV cache.
Unlike the above overload, this function returns the inference results directly instead of modifying an output parameter.
Note
This function is relevant for LLM models that use KV cache.
Parameters
input[in] — A vector of float pointers, where each pointer represents input data in NCHW or CHW format.shape[in] — A vector of vectors, where each inner vector specifies the shape of the corresponding input data.cache_size[in] — The number of tokens accumulated in the KV cache so far.sc[out] — A reference to a status code that will be updated to indicate whether the inference operation was successful or encountered an error.
Returns
A vector of float vectors containing the inference results.
NHWC uint8_t-to-float inference¶
Performs inference with input and output elements of type uint8_t in NHWC (batch N, height H, width W, channels C) or HWC format.
Note
The params (BatchParam) argument is not supported yet in these overloads. It is silently ignored, and inference is performed as if no BatchParam had been given. For BatchLLM inference, use the float input overloads instead.
| Member | Description |
|---|---|
infer ×14 |
infer¶
StatusCode infer(const std::vector<uint8_t *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape)
std::vector<std::vector<float>> infer(const std::vector<uint8_t *> &input, const std::vector<std::vector<int64_t>> &shape, StatusCode &sc)
StatusCode infer(const std::vector<NDArray<uint8_t>> &input, std::vector<NDArray<float>> &output, uint32_t cache_size)
std::vector<NDArray<float>> infer(const std::vector<NDArray<uint8_t>> &input, uint32_t cache_size, StatusCode &sc)
StatusCode infer(const std::vector<uint8_t *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size)
std::vector<std::vector<float>> infer(const std::vector<uint8_t *> &input, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size, StatusCode &sc)
StatusCode infer(const std::vector<NDArray<uint8_t>> &input, std::vector<NDArray<float>> &output, const std::vector<BatchParam> ¶ms)
std::vector<NDArray<float>> infer(const std::vector<NDArray<uint8_t>> &input, const std::vector<BatchParam> ¶ms, StatusCode &sc)
StatusCode infer(const std::vector<uint8_t *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape, const std::vector<BatchParam> ¶ms)
std::vector<std::vector<float>> infer(const std::vector<uint8_t *> &input, const std::vector<std::vector<int64_t>> &shape, const std::vector<BatchParam> ¶ms, StatusCode &sc)
NCHW uint8_t-to-float inference¶
Performs inference with input and output elements of type uint8_t in NCHW (batch N, channels C, height H, width W) or CHW format.
| Member | Description |
|---|---|
inferCHW ×10 |
inferCHW¶
StatusCode inferCHW(const std::vector<NDArray<uint8_t>> &input, std::vector<NDArray<float>> &output)
StatusCode inferCHW(const std::vector<uint8_t *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape)
std::vector<std::vector<float>> inferCHW(const std::vector<uint8_t *> &input, const std::vector<std::vector<int64_t>> &shape, StatusCode &sc)
StatusCode inferCHW(const std::vector<NDArray<uint8_t>> &input, std::vector<NDArray<float>> &output, uint32_t cache_size)
std::vector<NDArray<float>> inferCHW(const std::vector<NDArray<uint8_t>> &input, uint32_t cache_size, StatusCode &sc)
StatusCode inferCHW(const std::vector<uint8_t *> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size)
std::vector<std::vector<float>> inferCHW(const std::vector<uint8_t *> &input, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size, StatusCode &sc)
NHWC int8_t-to-int8_t inference¶
Performs inference with input and output elements of type int8_t in NHWC (batch N, height H, width W, channels C) or HWC format.
Using these inference APIs requires manual scaling (quantization) of float values to int8_t for input and int8_t to float for output.
Note
These APIs are intended for advanced use rather than typical usage.
Note
The params (BatchParam) argument is not supported yet in these overloads. It is silently ignored, and inference is performed as if no BatchParam had been given. For BatchLLM inference, use the float input overloads instead.
| Member | Description |
|---|---|
infer ×14 |
infer¶
StatusCode infer(const std::vector<int8_t *> &input, std::vector<std::vector<int8_t>> &output, const std::vector<std::vector<int64_t>> &shape)
std::vector<std::vector<int8_t>> infer(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, StatusCode &sc)
StatusCode infer(const std::vector<NDArray<int8_t>> &input, std::vector<NDArray<int8_t>> &output, uint32_t cache_size)
std::vector<NDArray<int8_t>> infer(const std::vector<NDArray<int8_t>> &input, uint32_t cache_size, StatusCode &sc)
StatusCode infer(const std::vector<int8_t *> &input, std::vector<std::vector<int8_t>> &output, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size)
std::vector<std::vector<int8_t>> infer(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size, StatusCode &sc)
StatusCode infer(const std::vector<NDArray<int8_t>> &input, std::vector<NDArray<int8_t>> &output, const std::vector<BatchParam> ¶ms)
std::vector<NDArray<int8_t>> infer(const std::vector<NDArray<int8_t>> &input, const std::vector<BatchParam> ¶ms, StatusCode &sc)
StatusCode infer(const std::vector<int8_t *> &input, std::vector<std::vector<int8_t>> &output, const std::vector<std::vector<int64_t>> &shape, const std::vector<BatchParam> ¶ms)
std::vector<std::vector<int8_t>> infer(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, const std::vector<BatchParam> ¶ms, StatusCode &sc)
NCHW int8_t-to-int8_t inference¶
Performs inference with input and output elements of type int8_t in NCHW (batch N, channels C, height H, width W) or CHW format.
Using these inference APIs requires manual scaling (quantization) of float values to int8_t for input and int8_t to float for output.
Note
These APIs are intended for advanced use rather than typical usage.
| Member | Description |
|---|---|
inferCHW ×10 |
inferCHW¶
StatusCode inferCHW(const std::vector<NDArray<int8_t>> &input, std::vector<NDArray<int8_t>> &output)
StatusCode inferCHW(const std::vector<int8_t *> &input, std::vector<std::vector<int8_t>> &output, const std::vector<std::vector<int64_t>> &shape)
std::vector<std::vector<int8_t>> inferCHW(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, StatusCode &sc)
StatusCode inferCHW(const std::vector<NDArray<int8_t>> &input, std::vector<NDArray<int8_t>> &output, uint32_t cache_size)
std::vector<NDArray<int8_t>> inferCHW(const std::vector<NDArray<int8_t>> &input, uint32_t cache_size, StatusCode &sc)
StatusCode inferCHW(const std::vector<int8_t *> &input, std::vector<std::vector<int8_t>> &output, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size)
std::vector<std::vector<int8_t>> inferCHW(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size, StatusCode &sc)
NHWC int8_t-to-float inference¶
Performs inference with input and output elements of type int8_t in NHWC (batch N, height H, width W, channels C) or HWC format.
Using these inference APIs requires manual scaling (quantization) of float values to int8_t for input.
Note
These APIs are intended for advanced use rather than typical usage.
| Member | Description |
|---|---|
inferToFloat ×5 |
inferToFloat¶
std::vector<std::vector<float>> inferToFloat(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, StatusCode &sc)
std::vector<NDArray<float>> inferToFloat(const std::vector<NDArray<int8_t>> &input, uint32_t cache_size, StatusCode &sc)
std::vector<std::vector<float>> inferToFloat(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size, StatusCode &sc)
NCHW int8_t-to-float inference¶
Performs inference with input and output elements of type int8_t in NCHW (batch N, channels C, height H, width W) or CHW format.
Using these inference APIs requires manual scaling (quantization) of float values to int8_t for input.
Note
These APIs are intended for advanced use rather than typical usage.
| Member | Description |
|---|---|
inferCHWToFloat ×5 |
inferCHWToFloat¶
std::vector<NDArray<float>> inferCHWToFloat(const std::vector<NDArray<int8_t>> &input, StatusCode &sc)
std::vector<std::vector<float>> inferCHWToFloat(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, StatusCode &sc)
std::vector<NDArray<float>> inferCHWToFloat(const std::vector<NDArray<int8_t>> &input, uint32_t cache_size, StatusCode &sc)
std::vector<std::vector<float>> inferCHWToFloat(const std::vector<int8_t *> &input, const std::vector<std::vector<int64_t>> &shape, uint32_t cache_size, StatusCode &sc)
NHWC Buffer-to-Buffer inference¶
Performs inference using input and output elements in the NPU’s internal data type. The inference operates on buffers allocated via the following APIs:
Model::acquireInputBufferModel::acquireOutputBufferModel::acquireInputBuffersModel::acquireOutputBuffersModelVariantHandle::acquireInputBufferModelVariantHandle::acquireOutputBufferModelVariantHandle::acquireInputBuffersModelVariantHandle::acquireOutputBuffers
Additionally, Model::repositionInputs, Model::repositionOutputs, ModelVariantHandle::repositionInputs and ModelVariantHandle::repositionOutputs must be used properly.
Note
These APIs are intended for advanced use rather than typical usage.
| Member | Description |
|---|---|
inferBuffer ×2 |
inferBuffer¶
StatusCode inferBuffer(const std::vector<Buffer> &input, std::vector<Buffer> &output, const std::vector<std::vector<int64_t>> &shape={}, uint32_t cache_size=0)
StatusCode inferBuffer(const std::vector<std::vector<Buffer>> &input, std::vector<std::vector<Buffer>> &output, const std::vector<std::vector<int64_t>> &shape={}, uint32_t cache_size=0)
NHWC Buffer-to-float inference¶
Performs inference using input and output elements in the NPU’s internal data type. The inference operates on buffers allocated via the following APIs:
Model::acquireInputBufferModel::acquireInputBuffersModelVariantHandle::acquireInputBufferModelVariantHandle::acquireInputBuffers
Additionally, Model::repositionInputs and ModelVariantHandle::repositionInputs must be used properly.
Note
These APIs are intended for advanced use rather than typical usage.
| Member | Description |
|---|---|
inferBufferToFloat ×4 |
inferBufferToFloat¶
StatusCode inferBufferToFloat(const std::vector<Buffer> &input, std::vector<NDArray<float>> &output, const std::vector<std::vector<int64_t>> &shape={}, uint32_t cache_size=0)
StatusCode inferBufferToFloat(const std::vector<std::vector<Buffer>> &input, std::vector<NDArray<float>> &output, const std::vector<std::vector<int64_t>> &shape={}, uint32_t cache_size=0)
StatusCode inferBufferToFloat(const std::vector<Buffer> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape={}, uint32_t cache_size=0)
StatusCode inferBufferToFloat(const std::vector<std::vector<Buffer>> &input, std::vector<std::vector<float>> &output, const std::vector<std::vector<int64_t>> &shape={}, uint32_t cache_size=0)
Asynchronous Inference¶
Performs inference asynchronously.
To use asynchronous inference, the model must be created using a ModelConfig object with the async pipeline configured to be enabled. This is done by calling ModelConfig::setAsyncPipelineEnabled(true) before passing the configuration to Model::create.
Example:
using namespace mobilint;
ModelConfig mc;
// Enables support for `inferAsync` and `inferAsyncCHW`
mc.setAsyncPipelineEnabled(true);
StatusCode sc;
std::unique_ptr<Model> model = Model::create("resnet50.mxq", mc, sc);
if (!sc) {
// Handle error appropriately
}
// Now `inferAsync` can be called.
Future future = model->inferAsync(input, sc);
Note
Functions in the inferAsync family (inferAsync, inferAsyncCHW, inferAsyncToFloat, inferAsyncCHWToFloat) typically return immediately. However, they may block if the input queue in the qbruntime is full.
Note
For all functions in the inferAsync family (inferAsync, inferAsyncCHW, inferAsyncToFloat, inferAsyncCHWToFloat), the data provided through the input parameter must remain unmodified until the asynchronous inference has completed. Modifying this data during execution may result in invalid results.
Note
Currently, only CNN-based models are supported, as asynchronous execution is particularly effective for this type of workload.
Note
Limitations:
- RNN/LSTM and LLM models are not supported yet.
- Models requiring CPU offloading are not supported yet.
- Currently, only single-batch inference is supported (i.e., N = 1).
- Currently, Buffer inference is not supported. The following types are supported in the synchronous API for advanced use cases, but are not yet available for asynchronous inference:
| Member | Description |
|---|---|
inferAsync ×3 |
Initiates asynchronous inference with input in NHWC (batch N, height H, width W, channels C) or HWC format. |
inferAsyncCHW ×3 |
Initiates asynchronous inference with input in NCHW (batch N, channels C, height H, width W) or CHW format. |
inferAsyncToFloat |
This overload supports int8_t-to-float asynchronous inference. |
inferAsyncCHWToFloat |
This overload supports int8_t-to-float asynchronous inference. |
inferAsync¶
Initiates asynchronous inference with input in NHWC (batch N, height H, width W, channels C) or HWC format.
Parameters
input[in] — A vector ofNDArray<float>. Each NDArray must be in NHWC or HWC format.sc[out] — A reference to a status code that will be updated to indicate whether the asynchronous inference request was successfully initiated or encountered an error.
Returns
A future that can be used to retrieve the inference result.
This overload supports int8_t-to-int8_t asynchronous inference.
Parameters
input[in] — A vector ofNDArray<int8_t>. Each NDArray must be in NHWC or HWC format.sc[out] — A reference to a status code that will be updated to indicate whether the asynchronous inference request was successfully initiated or encountered an error.
Returns
A future that can be used to retrieve the inference result.
This overload supports uint8_t-to-float asynchronous inference.
Parameters
input[in] — A vector ofNDArray<uint8_t>. Each NDArray must be in NHWC or HWC format.sc[out] — A reference to a status code that will be updated to indicate whether the asynchronous inference request was successfully initiated or encountered an error.
Returns
A future that can be used to retrieve the inference result.
inferAsyncCHW¶
Initiates asynchronous inference with input in NCHW (batch N, channels C, height H, width W) or CHW format.
Parameters
input[in] — A vector ofNDArray<float>. Each NDArray must be in NCHW or CHW format.sc[out] — A reference to a status code that will be updated to indicate whether the asynchronous inference request was successfully initiated or encountered an error.
Returns
A future that can be used to retrieve the inference result.
This overload supports int8_t-to-int8_t asynchronous inference.
Parameters
input[in] — A vector ofNDArray<int8_t>. Each NDArray must be in NCHW or CHW format.sc[out] — A reference to a status code that will be updated to indicate whether the asynchronous inference request was successfully initiated or encountered an error.
Returns
A future that can be used to retrieve the inference result.
This overload supports uint8_t-to-float asynchronous inference.
Parameters
input[in] — A vector ofNDArray<uint8_t>. Each NDArray must be in NCHW or CHW format.sc[out] — A reference to a status code that will be updated to indicate whether the asynchronous inference request was successfully initiated or encountered an error.
Returns
A future that can be used to retrieve the inference result.
inferAsyncToFloat¶
This overload supports int8_t-to-float asynchronous inference.
Parameters
input[in] — A vector ofNDArray<int8_t>. Each NDArray must be in NHWC or HWC format.sc[out] — A reference to a status code that will be updated to indicate whether the asynchronous inference request was successfully initiated or encountered an error.
Returns
A future that can be used to retrieve the inference result.
inferAsyncCHWToFloat¶
This overload supports int8_t-to-float asynchronous inference.
Parameters
input[in] — A vector ofNDArray<int8_t>. Each NDArray must be in NCHW or CHW format.sc[out] — A reference to a status code that will be updated to indicate whether the asynchronous inference request was successfully initiated or encountered an error.
Returns
A future that can be used to retrieve the inference result.
Buffer Management APIs¶
These APIs are required when calling Model::inferBuffer or Model::inferBufferToFloat.
Buffers are acquired using:
acquireInputBufferacquireOutputBuffer
Any acquired buffer must be released using:
releaseBufferreleaseBuffers
Repositioning is handled by:
repositionInputsrepositionOutputs
Note
These APIs are intended for advanced use rather than typical usage.
| Member | Description |
|---|---|
acquireInputBuffer |
|
acquireOutputBuffer |
|
acquireInputBuffers |
|
acquireOutputBuffers |
|
releaseBuffer |
|
releaseBuffers |
|
repositionInputs ×4 |
|
repositionOutputs ×4 |
acquireInputBuffer¶
acquireOutputBuffer¶
acquireInputBuffers¶
std::vector<std::vector<Buffer>> acquireInputBuffers(const int batch_size, const std::vector<std::vector<int>> &seqlens={}) const
acquireOutputBuffers¶
std::vector<std::vector<Buffer>> acquireOutputBuffers(const int batch_size, const std::vector<std::vector<int>> &seqlens={}) const
releaseBuffer¶
releaseBuffers¶
repositionInputs¶
StatusCode repositionInputs(const std::vector<float *> &input, std::vector<Buffer> &input_buf, const std::vector<std::vector<int>> &seqlens={}) const
StatusCode repositionInputs(const std::vector<uint8_t *> &input, std::vector<Buffer> &input_buf, const std::vector<std::vector<int>> &seqlens={}) const
StatusCode repositionInputs(const std::vector<float *> &input, std::vector<std::vector<Buffer>> &input_buf, const std::vector<std::vector<int>> &seqlens={}) const
StatusCode repositionInputs(const std::vector<uint8_t *> &input, std::vector<std::vector<Buffer>> &input_buf, const std::vector<std::vector<int>> &seqlens={}) const
repositionOutputs¶
StatusCode repositionOutputs(const std::vector<Buffer> &output_buf, std::vector<float *> &output, const std::vector<std::vector<int>> &seqlens={}) const
StatusCode repositionOutputs(const std::vector<Buffer> &output_buf, std::vector<std::vector<float>> &output, const std::vector<std::vector<int>> &seqlens={}) const
StatusCode repositionOutputs(const std::vector<std::vector<Buffer>> &output_buf, std::vector<float *> &output, const std::vector<std::vector<int>> &seqlens={}) const
StatusCode repositionOutputs(const std::vector<std::vector<Buffer>> &output_buf, std::vector<std::vector<float>> &output, const std::vector<std::vector<int>> &seqlens={}) const
NPUData Management APIs¶
These APIs hand out NPUData objects, which represent a single model input or output tensor and own the storage backing it on the host and/or the NPU.
Warning
This API is in beta: it may still contain bugs, and its behavior may change in a future release.
Note
These APIs are intended for advanced use rather than typical usage.
| Member | Description |
|---|---|
acquireInputNPUData ×2 |
Acquires an NPUData for the model input at the given index. |
acquireOutputNPUData ×2 |
Acquires an NPUData for the model output at the given index. |
infer ×2 |
Performs inference using NPUData for both inputs and outputs. |
inferCHW ×2 |
acquireInputNPUData¶
NPUData acquireInputNPUData(const std::vector<int64_t> &shape, int idx, bool upload, StatusCode &sc)
Acquires an NPUData for the model input at the given index.
Parameters
shape[in] — The shape of the input tensor, in NHWC/HWC or NCHW/CHW layout.idx[in] — The index of the model input.upload[in] — If true, NPU memory is allocated and the data is placed on the NPU; otherwise the data is kept on the host (CPU).sc[out] — A reference to a status code that will be updated to indicate whether the operation was successful or encountered an error.
Returns
The acquired NPUData. Its contents are valid only when sc is OK.
NPUData acquireInputNPUData(const std::vector<int64_t> &shape, int idx, bool upload, DataType dtype, StatusCode &sc)
Acquires an NPUData with an explicit element type.
Same as the overload above, except that the element type is chosen by the caller instead of defaulting to the model's user-facing input type. Use this to hand the model data that is already quantized to the NPU-native type (typically DataType::Int8), which skips the host-side quantization the default (DataType::Float32 / DataType::Uint8) types perform.
Parameters
shape[in] — The shape of the input tensor, in NHWC/HWC or NCHW/CHW layout.idx[in] — The index of the model input.upload[in] — If true, NPU memory is allocated and the data is placed on the NPU; otherwise the data is kept on the host (CPU).dtype[in] — The element type of the data. Must be one of the types the model accepts for this input; otherwisescis set toModel_DtypeMismatched.sc[out] — A reference to a status code that will be updated to indicate whether the operation was successful or encountered an error.
Returns
The acquired NPUData. Its contents are valid only when sc is OK.
acquireOutputNPUData¶
NPUData acquireOutputNPUData(const std::vector<int64_t> &shape, int idx, bool upload, StatusCode &sc)
Acquires an NPUData for the model output at the given index.
Parameters
shape[in] — The shape of the output tensor, in NHWC/HWC or NCHW/CHW layout.idx[in] — The index of the model output.upload[in] — If true, NPU memory is allocated and the data is placed on the NPU; otherwise the data is kept on the host (CPU).sc[out] — A reference to a status code that will be updated to indicate whether the operation was successful or encountered an error.
Returns
The acquired NPUData. Its contents are valid only when sc is OK.
NPUData acquireOutputNPUData(const std::vector<int64_t> &shape, int idx, bool upload, DataType dtype, StatusCode &sc)
Acquires an NPUData with an explicit element type.
Same as the overload above, except that the element type is chosen by the caller instead of defaulting to the model's user-facing output type. Use this to receive the raw NPU-native output (typically DataType::Int8), which skips the host-side dequantization the default type performs.
Parameters
shape[in] — The shape of the output tensor, in NHWC/HWC or NCHW/CHW layout.idx[in] — The index of the model output.upload[in] — If true, NPU memory is allocated and the data is placed on the NPU; otherwise the data is kept on the host (CPU).dtype[in] — The element type of the data. Must be one of the types the model produces for this output; otherwisescis set toModel_DtypeMismatched.sc[out] — A reference to a status code that will be updated to indicate whether the operation was successful or encountered an error.
Returns
The acquired NPUData. Its contents are valid only when sc is OK.
infer¶
StatusCode infer(std::vector<NPUData> &inputs, std::vector<NPUData> &outputs, uint32_t cache_size=0)
Performs inference using NPUData for both inputs and outputs.
Every input and output NPUData must share the same residency: either all of them are on the host (acquired with upload = false) or all of them are already on the NPU (acquired with upload = true). A mixed set results in an error.
When all of them are on the NPU, no reposition or copy is performed; the model instructions are relocated for the given input/output physical-address combination (only the first time a given combination is seen) and the NPU runs directly. The results are left in the output NPU memory and can be retrieved afterwards.
When all of them are on the host, the inputs are additionally repositioned and uploaded to the NPU before execution, and the outputs are downloaded and repositioned back to the host afterwards.
Warning
This API is in beta: it may still contain bugs, and its behavior may change in a future release.
Note
These APIs are intended for advanced use rather than typical usage.
Note
Only single-NPU-op (non-CPU-offload), relocatable (MXQv7+) models are supported.
Note
infer requires every NPUData to have been acquired with an NHWC/HWC shape and inferCHW with an NCHW/CHW shape; a mismatch fails with Model_ShapeMismatched.
Parameters
inputs[in,out] — A vector of inputNPUData.outputs[in,out] — A vector of outputNPUData.cache_size[in] — The number of tokens accumulated in the KV cache so far.
Returns
A status code indicating whether the inference operation completed successfully or encountered an error.
Performs inference using NPUData inputs and returns the outputs.
Unlike the above overload, the output NPUData are acquired internally (sized from the matching model variant and given the same residency as the inputs) and returned directly instead of being supplied by the caller.
Warning
This API is in beta: it may still contain bugs, and its behavior may change in a future release.
Note
These APIs are intended for advanced use rather than typical usage.
Note
Only single-NPU-op (non-CPU-offload), relocatable (MXQv7+) models are supported.
Note
infer requires every NPUData to have been acquired with an NHWC/HWC shape and inferCHW with an NCHW/CHW shape; a mismatch fails with Model_ShapeMismatched.
Parameters
inputs[in,out] — A vector of inputNPUData.sc[out] — A reference to a status code that will be updated to indicate success or the specific error condition.cache_size[in] — The number of tokens accumulated in the KV cache so far.
Returns
A vector of output NPUData containing the inference results. Its contents are valid only when sc is OK.
inferCHW¶
StatusCode inferCHW(std::vector<NPUData> &inputs, std::vector<NPUData> &outputs, uint32_t cache_size=0)
KV Cache Management¶
Note
These APIs are used for LLM models that utilize KV cache.
| Member | Description |
|---|---|
dumpCacheMemory ×4 |
Dumps the KV cache memory into buffers. |
loadCacheMemory ×2 |
Loads the KV cache memory from buffers. |
dumpCacheMemoryByName ×2 |
Dumps a single KV cache memory into a buffer. |
dumpCacheMemoryByLayerHash ×2 |
Dumps a single KV cache memory into a buffer. |
loadCacheMemoryByName |
Loads a single KV cache memory from a buffer. |
loadCacheMemoryByLayerHash |
Loads a single KV cache memory from a buffer. |
resetFixedTailCacheMemory |
Reset the fixed tail KV cache memory. |
filterCacheTail |
Filter the tail of the KV cache memory. |
moveCacheTail |
Moves the tail of the KV cache memory to the end of the head. |
inferPinnedMemory ×2 |
Performs inference directly on pinned memory buffers. |
dumpCacheMemory¶
Dumps the KV cache memory into buffers.
Writes the current KV cache data into provided buffers.
Parameters
bufs[out] — A reference to vectors of byte vectors that will store the KV cache data.cache_id[in] — Index of target cache.
Returns
A status code indicating whether the dump operation was successful or if an error occurred.
Dumps the KV cache memory into buffers.
Writes the KV cache data into buffers and returns them.
Parameters
cache_id[in] — Index of target cache.sc[out] — A reference to a status code that will be updated to indicate whether the dump operation was successful or if an error occurred.
Returns
A vector of byte vectors containing the KV cache data.
Dumps KV cache memory to files in the specified directory.
Writes the KV cache data to binary files within the given directory. Each file is named using the format: cache_<layer_hash>.bin.
Parameters
cache_dir[in] — Path to the directory where KV cache files will be saved.cache_id[in] — Index of target cache.
Returns
A status code indicating whether the dump operation was successful or if an error occurred.
loadCacheMemory¶
Loads the KV cache memory from buffers.
Restores the KV cache from the provided buffers.
Parameters
bufs[in] — A reference to a vector of byte vectors containing the KV cache data.cache_id[in] — Index of target cache.
Returns
A status code indicating whether the load operation was successful or if an error occurred.
Loads the KV cache memory from files in the specified directory.
Reads KV cache data from files within the given directory and restores them. Each file is named using the format: cache_<layer_hash>.bin.
Parameters
cache_dir[in] — Path to the directory where KV cache files are saved.cache_id[in] — Index of target cache.
Returns
A status code indicating whether the load operation was successful or if an error occurred.
dumpCacheMemoryByName¶
Dumps a single KV cache memory into a buffer.
Writes the current data of the KV cache whose CacheInfo::name matches name into the provided buffer.
Parameters
buf[out] — A reference to a byte vector that will store the KV cache data. If it is empty, it is resized to the size of the target KV cache.name[in] — Name of the target KV cache. SeeCacheInfo::name.cache_id[in] — Index of target cache.
Returns
A status code indicating whether the dump operation was successful or if an error occurred.
Dumps a single KV cache memory into a buffer.
Writes the data of the KV cache whose CacheInfo::name matches name into a buffer and returns it.
Parameters
name[in] — Name of the target KV cache. SeeCacheInfo::name.sc[out] — A reference to a status code that will be updated to indicate whether the dump operation was successful or if an error occurred.cache_id[in] — Index of target cache.
Returns
A byte vector containing the KV cache data.
dumpCacheMemoryByLayerHash¶
StatusCode dumpCacheMemoryByLayerHash(std::vector<int8_t> &buf, const std::string &layer_hash, int cache_id=0)
Dumps a single KV cache memory into a buffer.
Writes the current data of the KV cache whose CacheInfo::layer_hash matches layer_hash into the provided buffer.
Parameters
buf[out] — A reference to a byte vector that will store the KV cache data. If it is empty, it is resized to the size of the target KV cache.layer_hash[in] — Layer hash of the target KV cache. SeeCacheInfo::layer_hash.cache_id[in] — Index of target cache.
Returns
A status code indicating whether the dump operation was successful or if an error occurred.
std::vector<int8_t> dumpCacheMemoryByLayerHash(const std::string &layer_hash, StatusCode &sc, int cache_id=0)
Dumps a single KV cache memory into a buffer.
Writes the data of the KV cache whose CacheInfo::layer_hash matches layer_hash into a buffer and returns it.
Parameters
layer_hash[in] — Layer hash of the target KV cache. SeeCacheInfo::layer_hash.sc[out] — A reference to a status code that will be updated to indicate whether the dump operation was successful or if an error occurred.cache_id[in] — Index of target cache.
Returns
A byte vector containing the KV cache data.
loadCacheMemoryByName¶
StatusCode loadCacheMemoryByName(const std::vector<int8_t> &buf, const std::string &name, int cache_id=0)
Loads a single KV cache memory from a buffer.
Restores only the KV cache whose CacheInfo::name matches name from the provided buffer. The other KV caches are left untouched.
Parameters
buf[in] — A byte vector containing the KV cache data. Its size must be equal to the size of the target KV cache.name[in] — Name of the target KV cache. SeeCacheInfo::name.cache_id[in] — Index of target cache.
Returns
A status code indicating whether the load operation was successful or if an error occurred.
loadCacheMemoryByLayerHash¶
StatusCode loadCacheMemoryByLayerHash(const std::vector<int8_t> &buf, const std::string &layer_hash, int cache_id=0)
Loads a single KV cache memory from a buffer.
Restores only the KV cache whose CacheInfo::layer_hash matches layer_hash from the provided buffer. The other KV caches are left untouched.
Parameters
buf[in] — A byte vector containing the KV cache data. Its size must be equal to the size of the target KV cache.layer_hash[in] — Layer hash of the target KV cache. SeeCacheInfo::layer_hash.cache_id[in] — Index of target cache.
Returns
A status code indicating whether the load operation was successful or if an error occurred.
resetFixedTailCacheMemory¶
Reset the fixed tail KV cache memory.
Fills the entire memory region of every KV cache whose CacheType is CacheType::FixedTailCache with zeros. The caches of the other types are left untouched.
Note
This does not update the cache size tracked by the caller; the caller must reset its own cache_size to 0 after this call.
Parameters
cache_id[in] — Index of target cache.
Returns
A status code indicating whether the reset operation was successful or if an error occurred.
filterCacheTail¶
Filter the tail of the KV cache memory.
Retains the desired caches in the tail of the KV cache memory, excludes the others, and shifts the remaining caches forward.
Parameters
cache_size[in] — The number of tokens accumulated in the KV cache so far.tail_size[in] — The tail size of the KV cache to filter (<=32).mask[in] — A mask indicating tokens to retain or exclude at the tail of the KV cache.sc[out] — A status code indicating the outcome of the tail filtering.
Returns
New cache size after tail filtering.
moveCacheTail¶
Moves the tail of the KV cache memory to the end of the head.
Slice the tail of the KV cache memory up to the specified size and moves it to the designated cache position.
Parameters
num_head[in] — The size of the KV cache head where the tail is appended.num_tail[in] — The size of the KV cache tail to be moved.cache_size[in] — The total number of tokens accumulated in the KV cache so far.sc[out] — A status code indicating the result of the tail move.
Returns
The updated cache size after moving the tail.
inferPinnedMemory¶
template <typename T>
std::vector<PinnedMemory<float>> inferPinnedMemory(const std::vector<PinnedMemory<T>> &inputs, uint32_t cache_size, StatusCode &sc)
Performs inference directly on pinned memory buffers.
This overload differs from the one below in that it directly returns the output pinned memory buffers instead of modifying an output parameter. The returned buffers are newly allocated.
Runs inference using NPU-accessible pinned buffers as the input storage and freshly allocated pinned buffers as the output storage, avoiding host-to-device copies of the I/O tensors.
Note
This is an experimental API and is only supported on REGULUS device. It is not supported for models that use CPU offload.
Note
The inputs buffers must be pre-allocated via Accelerator::allocatePinnedMemory and must be available (see PinnedMemory::isAvailable) before calling this function.
Note
Pinned inference on a given model must be called serially. Concurrent pinned inference calls on the same model are not supported.
Template parameters
T— The element type of the input buffers (uint8_torfloat).
Parameters
inputs[in] — A vector of pinned memory buffers containing the input data.cache_size[in] — The size of the cache to use for inference.sc[out] — A reference to a status code that will be updated to indicate whether the inference was successful or if an error occurred.
Returns
A vector of newly allocated pinned memory buffers containing the inference results.
template <typename T>
StatusCode inferPinnedMemory(const std::vector<PinnedMemory<T>> &inputs, std::vector<PinnedMemory<float>> &outputs, uint32_t cache_size)
Performs inference directly on pinned memory buffers.
Runs inference using NPU-accessible pinned buffers as the input and output storage, avoiding host-to-device copies of the I/O tensors.
Note
This is an experimental API and is only supported on REGULUS device. It is not supported for models that use CPU offload.
Note
The inputs and outputs buffers must be pre-allocated via Accelerator::allocatePinnedMemory and must be available (see PinnedMemory::isAvailable) before calling this function.
Note
If outputs is empty (i.e. its size is 0), the qbruntime allocates the required output pinned memory buffers internally and populates outputs with them. Otherwise, outputs must be pre-allocated buffers whose count and shapes match the model's output; a mismatch results in an error.
Note
Pinned inference on a given model must be called serially. Concurrent pinned inference calls on the same model are not supported.
Template parameters
T— The element type of the input buffers (uint8_torfloat).
Parameters
inputs[in] — A vector of pinned memory buffers containing the input data.outputs[out] — A vector of pinned memory buffers to store the output data. If empty, output buffers are allocated internally and returned through this parameter.cache_size[in] — The size of the cache to use for inference.
Returns
A status code indicating whether the inference was successful or if an error occurred.
Deprecated APIs¶
Note
These APIs are deprecated and should not be used.
| Member | Description |
|---|---|
infer ×2 |
infer¶
StatusCode infer(const std::vector<float *> &input, std::vector<std::vector<float>> &output, int batch_size)
Deprecated
Use infer(input, output, shape) instead.
std::vector<std::vector<float>> infer(const std::vector<float *> &input, int batch_size, StatusCode &sc)
Deprecated
Use infer(input, shape, sc) instead.
Static functions¶
| Member | Description |
|---|---|
create ×2 |
Creates a Model object from the specified MXQ model file. |
create¶
Creates a Model object from the specified MXQ model file.
Parses the MXQ file and constructs a Model object. The model is initialized in single-core mode with all NPU local cores included.
Note
The created Model object must be launched before performing inference. See Model::launch for more details.
Parameters
mxq_path[in] — The path to the MXQ model file.sc[out] — A reference to a status code that will be updated to indicate whether the model was successfully created or if an error occurred.
Returns
A unique pointer to the created Model object.
static std::unique_ptr<Model> create(const std::string &mxq_path, const ModelConfig &config, StatusCode &sc)
Creates a Model object from the specified MXQ model file and configuration.
Parses the MXQ file and constructs a Model object using the provided configuration, initializing the model with the given settings.
Note
The created Model object must be launched before performing inference. See Model::launch for more details.
Parameters
mxq_path[in] — The path to the MXQ model file.config[in] — The configuration settings to initialize the Model.sc[out] — A reference to a status code that will be updated to indicate whether the model was successfully created or if an error occurred.
Returns
A unique pointer to the created Model object.
Member functions¶
| Member | Description |
|---|---|
Model ×2 |
|
operator= ×2 |
|
launch |
Launches the model on the specified Accelerator, which represents the actual NPU. |
dispose |
Disposes of the model loaded onto the NPU. |
getCoreMode |
Retrieves the core mode of the model. |
isTarget |
Checks if the NPU core specified by CoreId is the target of the model. In other words, whether the model is configured to use the given NPU core. |
getTargetCores |
Returns the NPU cores the model is configured to use. |
inferSpeedrun |
Development-only API for measuring pure NPU inference speed. |
getNumModelVariants |
Returns the total number of model variants available in this model. |
getModelVariantHandle |
Retrieves a handle to the specified model variant. |
getModelInputShape |
Returns the input shape of the model. |
getModelOutputShape |
Returns the output shape of the model. |
getInputBufferInfo |
Returns the input buffer information for the model. |
getOutputBufferInfo |
Returns the output buffer information of the model. |
getInputScale |
Returns the input quantization scale(s) of the model. |
getOutputScale |
Returns the output quantization scale(s) of the model. |
getModelInputDataType |
Returns a data type for model inputs. |
getModelOutputDataType |
Returns a data type for model outputs. |
getIdentifier |
Returns the model's unique identifier. |
getModelPath |
Returns the path to the MXQ model file associated with the Model. |
getDeviceNames |
Returns the supported target device name(s) this model can run on. |
getCacheInfos |
Returns informations of KV-cache of the model. |
Model¶
operator=¶
launch¶
Launches the model on the specified Accelerator, which represents the actual NPU.
Parameters
acc[in] — The accelerator on which to launch the model.
Returns
A status code indicating whether the model was successfully launched or if an error occurred.
dispose¶
Disposes of the model loaded onto the NPU.
Releases any resources associated with the model on the NPU.
Returns
A status code indicating whether the disposal was successful or if an error occurred.
getCoreMode¶
Retrieves the core mode of the model.
Returns
The CoreMode of the model.
isTarget¶
Checks if the NPU core specified by CoreId is the target of the model. In other words, whether the model is configured to use the given NPU core.
Parameters
core_id[in] — The CoreId to check.
Returns
True if the model is configured to use the specified CoreId, false otherwise.
getTargetCores¶
Returns the NPU cores the model is configured to use.
Returns
A vector of CoreIds representing the target NPU cores.
inferSpeedrun¶
Development-only API for measuring pure NPU inference speed.
Runs NPU inference without uploading inputs and without retrieving outputs.
Note
For models with variable length inputs, this runs based on a sequence length of 1.
Note
For models using a kv-cache, this runs based on a cache size of 0.
Parameters
variant_idx[in] — Index of model variant to run
Returns
A status code indicating the result.
getNumModelVariants¶
Returns the total number of model variants available in this model.
The variant_idx parameter passed to Model::getModelVariantHandle must be in the range [0, return value of this function).
Returns
The total number of model variants.
getModelVariantHandle¶
Retrieves a handle to the specified model variant.
Use the returned ModelVariantHandle to query details such as input and output shapes for the selected variant.
Parameters
variant_idx[in] — Index of the model variant to retrieve. Must be in the range [0, getNumModelVariants()).sc[out] — A reference to a StatusCode variable that will be updated to indicate success or failure.
Returns
A unique pointer to the corresponding ModelVariantHandle if successful; otherwise, nullptr.
getModelInputShape¶
Returns the input shape of the model.
Returns
A reference to the input shape of the model.
getModelOutputShape¶
Returns the output shape of the model.
Returns
A reference to the output shape of the model.
getInputBufferInfo¶
Returns the input buffer information for the model.
Returns
A reference to a vector of input buffer information.
getOutputBufferInfo¶
Returns the output buffer information of the model.
Returns
A reference to a vector of output buffer information.
getInputScale¶
Returns the input quantization scale(s) of the model.
Returns
A vector of input scales.
getOutputScale¶
Returns the output quantization scale(s) of the model.
Returns
A vector of output scales.
getModelInputDataType¶
Returns a data type for model inputs.
Returns
An input data type.
getModelOutputDataType¶
Returns a data type for model outputs.
Returns
An output data type.
getIdentifier¶
Returns the model's unique identifier.
This identifier distinguishes multiple models within a single user program. It is assigned incrementally, starting from 0 (e.g., 0, 1, 2, 3, ...).
Returns
The model identifier.
getModelPath¶
Returns the path to the MXQ model file associated with the Model.
Returns
The MXQ file path.
getDeviceNames¶
Returns the supported target device name(s) this model can run on.
Returns target device names from the model's target device in the MXQ file. A single model may map to multiple names (e.g. a REGULUS model returns both SoC and USB devices). Any of the returned names can be passed to Accelerator::create.
Returns
A vector of supported target device names.
getCacheInfos¶
Returns informations of KV-cache of the model.
Returns
A vector of CacheInfo objects.