MachGen / open-vc-attentionView on GitHub
MachGen's VC attention (https://arxiv.org/html/2609.15810v1) implementation. 2x faster (compared to 1.47-1.6x in original paper) FP8 attention on Blackwell than BF16 FlashAttention for long sequence video inference.
☆15Oct 12, 2026Updated 3 weeks ago

Alternatives and similar repositories for open-vc-attention

Users that are interested in open-vc-attention are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.

Sorting:

Are these results useful?