https://pypi.org/project/fold-attention/ Recent updates to the Python Package Index for fold-attention en https://pypi.org/project/fold-attention/0.1.1/ FoldAttention declares softmax’s reference before execution, making attention additive and enabling up to 3.09× faster H100 decode and deterministic backward faster than nondeterministic kernels. srimanachanta@gmail.com Sun, 27 Sep 2026 01:28:03 GMT https://pypi.org/project/fold-attention/0.1.0/ FoldAttention declares softmax’s reference before execution, making attention additive and enabling up to 3.09× faster H100 decode and deterministic backward faster than nondeterministic kernels. srimanachanta@gmail.com Sun, 27 Sep 2026 01:17:23 GMT