In an iteration, each GPU sends a segment to the next GPU:

– GPU1 sends a₁ to GPU2, where it is added to b₁
– GPU2 sends b₂ to GPU3, where it is added to c₂
– GPU3 sends c₃ to GPU4, where it is added to d₃
– GPU4 sends d₄ to GPU1, where it is added to a₄

Check this 👇 pic.twitter.com/aMc08xJwhp

— Akshay 🚀 (@akshay_pachaar) 17 août 2025

In an iteration, each GPU sends a segment to the next GPU: – GPU1 sends a₁ to GPU2, where it is added to b₁
– GPU2 sends b₂ to GPU3, where it is added to c₂
– GPU3 sends c₃ to GPU4, where it is added to d₃
– GPU4 sends d₄ to GPU1, where it is added to a₄ Check this

→ View original post on X — @akshay_pachaar