I attached all 3 algorithms 1F1B (1 forward 1 backward), ZB1P (zero bubble pipeline parallelism) and DualPipe as a picture here: <a href="https://x.com/danielhanchen/status/1894937006352031832" rel="nofollow">https://x.com/danielhanchen/status/1894937006352031832</a> for those interested :)
I hope all the open sources Deepseek is doing encourages American labs to do more of the same. Surely they'll realize their momentum is more of a moat than their tech at any one point in time.