# TorchDynamo Update 4: LazyTensor & nvFuser Experiments

**URL:** <https://dev-discuss.pytorch.org/t/torchdynamo-update-4-lazytensor-nvfuser-experiments/496>\
**Category:** compiler\
**Created:** [February 3, 2022, 1:35am UTC](https://dev-discuss.pytorch.org/t/torchdynamo-update-4-lazytensor-nvfuser-experiments/496 "2022-02-03T01:35:20Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![jansel](https://yyz2.discourse-cdn.com/flex036/user_avatar/dev-discuss.pytorch.org/jansel/32/74_2.png) [@jansel](https://dev-discuss.pytorch.org/u/jansel)\
**Post date:** [February 25, 2022, 5:09am UTC](https://dev-discuss.pytorch.org/t/torchdynamo-update-4-lazytensor-nvfuser-experiments/496/3 "2022-02-25T05:09:33Z")

</div>

@cicirori we have not done so yet, but are thinking about an XLA backend for TorchDynamo, so we could compare with LTC+XLA and see if these same results hold.

---

_[View the full topic](https://dev-discuss.pytorch.org/t/torchdynamo-update-4-lazytensor-nvfuser-experiments/496)._