Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Torch-Tensorrt Integration with LightningModule (#20808)
* feat: add `to_tensorrt` in the `LightningModule`. * refactor: fix `to_tensorrt` impl * test: add test_torch_tensorrt.py * add dependency in test requirement * limit the torch-tensorrt condition again * update tensorrt version * update tensorrt source * update test.txt * ci: add extra-index * ci: use find-links instead. works on my computer... * fix: fix bug in torch-tensorrt 2.8.0 * add find links in ci test. * chlog --------- Co-authored-by: Jirka Borovec <[email protected]> Co-authored-by: Jirka B <[email protected]>
G
GdoongMathew committed
39e24f5d37ba6421019254920967fc7e2d4c7ed6
Parent: 3ed9d4e
Committed by GitHub <[email protected]>
on 8/15/2025, 8:50:55 AM