teith

Results 5 comments of teith

[Feature]: Non-decaying laser

+1

Possible support for stable-cascade

up

Lower-than-Expected Performance Improvement with INT8 Quantization in TensorRT 10.0 on A100 GPU

Hi, @azhurkevich! > Do you think we can ask you to post your TRT engine/send it to us for analysis? As well as env you are using for repro Here...

Lower-than-Expected Performance Improvement with INT8 Quantization in TensorRT 10.0 on A100 GPU

Hi, @jingyu-ml > What node are you using? Are you using AWS or GCP or your own machine? I've used A100 40GB on Lambda Cloud

Lower-than-Expected Performance Improvement with INT8 Quantization in TensorRT 10.0 on A100 GPU

Hi, @jingyu-ml ! > is that possible for you to attach the int8 unet onnx file at somewhere? [model.onnx](https://trt-tests-public.s3.us-west-2.amazonaws.com/SDXL+Tests/SDXL-A100-TRT9.3-Quantization/unetxl-int8.l3.0.bs2.s30.c384.p0.4.a0.6.opt/model.onnx) [76d7dfc4-f8d8-11ee-a05c-0242ac120002](https://us-west-2.console.aws.amazon.com/s3/ap/988456508912/trt-tests-public/object?region=us-west-2&bucketType=general&prefix=SDXL+Tests/SDXL-A100-TRT9.3-Quantization/unetxl-int8.l3.0.bs2.s30.c384.p0.4.a0.6.opt/76d7dfc4-f8d8-11ee-a05c-0242ac120002)