[PT2.6] [Torch.complie] Implement AOTInuctor and torch.export on Intel GPU

intel / torch-xpu-ops

Apache License 2.0

30 stars 21 forks source link

[PT2.6] [Torch.complie] Implement AOTInuctor and torch.export on Intel GPU #1002

Open riverliuintel opened 1 month ago

riverliuintel commented 1 month ago

🚀 The feature, motivation and pitch

Implement the AOTInuctor and torch.export on Intel GPU Linux. It requires enabling the model binary store/load mechanism, support ABI netural calling and enabling feature UT.

Alternatives

No response

Additional context

No response