Search results

SegmentedPolynomial checkpoints saved on CUDA cannot be loaded on CPU due to missing __reduce__ on backend classes, preventing model portability

Status: Open.

#267 In NVIDIA/cuEquivariance;

· jchodera opened on Apr 8, 2026

SphericalHarmonics naive backend does not support torch.compile

Status: Open.

#265 In NVIDIA/cuEquivariance;

· rwkeane opened on Apr 8, 2026

Exception when using FullyConnectedTensorProductConv's second order derivatives

Status: Open.

#264 In NVIDIA/cuEquivariance;

· rwkeane opened on Apr 6, 2026

JIT Compilation issues?

Status: Open.

#259 In NVIDIA/cuEquivariance;

· RishabhDey opened on Mar 28, 2026

Out of shared memory on blackwell architectures with embedding size 128

bugSomething isn't workingSomething isn't working

Status: Open.

#244 In NVIDIA/cuEquivariance;

· psturmfels opened on Jan 28, 2026

torch.compile crashes when sizes are unbacked SymInts.

Status: Open.

#243 In NVIDIA/cuEquivariance;

· rwkeane opened on Jan 28, 2026

Typing issue for triangle_multiplicative_update when using cuEquivariance with openfold2

Status: Open.

#229 In NVIDIA/cuEquivariance;

· dxu16 opened on Jan 7, 2026

Conda packages for cuequivariance-ops-*

feature requestNew feature requestNew feature request

Status: Open.

#218 In NVIDIA/cuEquivariance;

· Emrys-Merlin opened on Dec 1, 2025

cuequivariance-ops-cu11 for aarch64

Status: Open.

#215 In NVIDIA/cuEquivariance;

· Swiftily opened on Nov 22, 2025

Bug: cuEquivariance-jax not support the B200 / B300 | cuda12

Status: Open.

#209 In NVIDIA/cuEquivariance;

· guyujun opened on Nov 10, 2025

Finer granularity of control over triangular multiplicative update

feature requestNew feature requestNew feature request

Status: Open.

#205 In NVIDIA/cuEquivariance;

· RuiWang1998 opened on Oct 23, 2025

Attention fwd/bwd issues with all zero masks

Status: Open.

#204 In NVIDIA/cuEquivariance;

· RuiWang1998 opened on Oct 23, 2025