PyTorch Edge

> We have created extensive documentation that provides more details about ExecuTorch’s architecture, its high-level components, example ML models running on ExecuTorch, and end-to-end tutorials for exporting and running a model on various hardware devices. We are excited to see all of the innovative use cases of ExecuTorch built by the community.

"extensive documentation" should be a link to said extensive documentation

(A seemingly nitpicky comment, but I think it's a simple yet important way of encouraging "customer" onboarding.)

There's a "learn more" link at the bottom, but I don't think users should have to weed through the whole page to get to the info they really want.

The "links as subject affinities" section in Every Page Is Page One by Mark Baker has a lot of thoughtful ideas along these lines.

basically a tensorflow lite for pytorch

Would be cool to show some benchmark numbers and the binary size