What's inference mode? Why doesn't my code run fast if I use no_grad or make sure requires_grad=False? How come inference mode is safe but AutoNonVariableTypeMode is not?Further reading:<ul><li>Inference mode RFC <a href="https://github.com/ailzhang/rfcs/blob/rfc0011/RFC-0011-InferenceMode.md" target="_blank">https://github.com/.../rfc0011/RFC-0011-InferenceMode.md</a></li><li>Inference mode docs for C++ frontend users <a href="https://github.com/pytorch/pytorch/blob/master/docs/cpp/source/notes/inference_mode.rst" target="_blank">https://github.com/.../cpp/source/notes/inference_mode.rst</a></li><li>Tracking issue for Python frontend support <a href="https://github.com/pytorch/pytorch/issues/56608" target="_blank">https://github.com/pytorch/pytorch/issues/56608</a></li></ul>

Just enough CUDA to be dangerous

The PyTorch Developer Podcast is a place for the PyTorch dev team to do bite sized (10-20 min) topics about all sorts of internal development topics in PyTorch.