We’re excited to announce the release of PyTorch 2.0! This version includes: ⚙️ 100% backward compatible 📦 Out of the box performance 📶 Significant speed improvements Learn more 👇 hubs.la/Q01H4L1t0

📚 We’re also including new libraries in the release of PyTorch 2.0, including those that are in-tree, and separate libraries including TorchAudio, TorchVision, and TorchText. Learn more 👇 pytorch.org/blog/new-libra…

@PyTorch As someone who has been a fan and user since 0.4 Thank you to everyone who has been behind all of this awesomeness and for making PyTorch the best DL framework for us 🙏

@PyTorch "100% backward compatible" - this is the way to do it. Super excited to try out torch.compile().

@PyTorch @RiversHaveWings 🥳🥳🥳

@PyTorch @ncooper57 Blog typo: Feature table, Performance Improvements column refers to Python 1.7 and 1.8 rather than 3.7 and 3.8

@PyTorch Whenever Pytorch release a new version, I find new interesting features. For Tensorflow, I found API changes and bugs

@PyTorch Any substantial gains on Apple Silicon?

@PyTorch What a week 💜

@PyTorch I must to try 👍👍

@PyTorch Does C++ version also has 2.0?

@PyTorch Awesome , “100% backward compatible”. Thanks

@PyTorch Congratulations on the release of PyTorch 2.0! Excited to try out the out of the box performance and significant speed improvements. Great work!

@PyTorch @DanielLenton1 This is new, I'm curious to know how can Ivy make sure it's compatibility with the latest versions everytime? :)

@PyTorch Can’t wait to give it a try.

@PyTorch Excited for the new developments, cant wait to try it out

@PyTorch Oooh🥹 at last, now py 3.11 can run with it

@PyTorch wow

@PyTorch Now let’s eat

@PyTorch rip tensorflow