Alex

Hi, I'm just curious about the speed of ResNet18 on mobile. Could you please give a hint? Thanks!

Context switching causes multi GPU idling

2

https://github.com/CanyonWind/Single-Path-One-Shot-NAS-MXNet/blob/df1c1ffdd150695745b5abd4002be330be5d13b0/oneshot_nas_blocks.py#L166 https://github.com/CanyonWind/Single-Path-One-Shot-NAS-MXNet/blob/df1c1ffdd150695745b5abd4002be330be5d13b0/oneshot_nas_blocks.py#L467-L470

Will you consider SELLING the full version for commercial usage?

1

Impressive work! If you do, please leave me a contact email. We are very interested in it. :)

Two of the ForeignNodes consumes 60% inference time, among 1000 nodes

3

## Description When running the tensorrt engine and do profiling (using `trtexec`), we found that two of the `ForeignNode` takes `60%` inference time. And the total number of nodes in...

Setting graph_mode to True causes no error enum: 900

6

Hi, when using AIT with graph mode to `False`, it works smoothly. But as the title says, setting it to `True` causes the below runtime error. This happens both on...

Any plan on supporting Flash Attention 2

1

[Flash attention 2](https://github.com/Dao-AILab/flash-attention) seems to be consistently faster than the first version flash attention. Wonder any plans on supporting it? Thanks

1) I downloaded the VGG_19 model from here: https://github.com/jcjohnson/neural-style/issues/202 2) Changed vgg_model and vgg_deploy path in make_flickr_dataset.py 3) Run make_flickr_dataset.py, it stucks at cnn = CNN(model=vgg_model, deploy=vgg_deploy, width=224, height=224) Any...

### Required prerequisites - [x] I have read the documentation . - [x] I have searched the [Issue Tracker](https://github.com/tile-ai/tilelang/issues) that this hasn't already been reported. (comment there if it has.)...