Model opt

Accelerated Computing DGX Spark / GB10 User Forum DGX Spark / GB10

978206970 June 3, 2026, 3:20am 1

Hi, I encountered an issue. I have tried quantizing the vit_base_patch16_224 model in INT4 and INT8. The problem is that when exporting to ONNX, the pre_process function in INT4QuantExporter can only match subgraphs like DequantizeLinear -> Reshape -> Transpose -> MatMul/Gemm. In fact, there are other subgraphs, such as DequantizeLinear -> Transpose -> MatMul/Gemm, which cannot be handled successfully. How can I fix this? Thanks.

Topic		Replies	Views
Int8 TensorCores for Jetson Jetson AGX Xavier tensorrt	6	1558	April 26, 2023
Post-Training Quantization (PTQ) for semantic segmentation model running on Jetson Orin NX Jetson Orin NX tensorrt	23	938	March 12, 2025
TensorRT generated QAT engine, why the engine is bigger than pretrained fp16 engine? TensorRT	3	1446	January 4, 2022
Trtexec conversion to int8 Jetson AGX Orin tensorrt	1	1440	October 11, 2023
Why while ONNX-TensorRT conversion with INT8 quantizations some layers are not quantized? TensorRT tensorrt , pytorch , onnx	12	3053	December 4, 2022
[Hugging Face transformer models + pytorch_quantization] PTQ quantization int8 is slower than fp16 TensorRT tensorrt , python , onnx , natural-language-processing-nlp	4	3238	January 6, 2022
How can we know we have convert the onnx to int8trt rather than Float32? TensorRT tensorrt	23	2260	June 14, 2021
Missing quantization data when converting TF2.x QAT => ONNX => TensorRT TensorRT tensorrt	14	2523	October 18, 2021
Convert int8-onnx model to trt engine? TensorRT onnx	6	1287	April 29, 2023
Convert onnx to tensorrt error on Jetson Xavier. Jetson AGX Xavier	5	1792	June 26, 2019

Model opt

Related topics