NVIDIA/cuEquivariance

View on GitHub

cuEquivariance is a math library that is a collective of low-level primitives and tensor ops to accelerate widely-used models, like DiffDock, MACE, Allegro and NEQUIP, based on equivariant neural networks. Also includes kernels for accelerated structure prediction.

Stars: 433
Forks: 42
Open beginner issues: 0
Indexed issues: 21

Dominant language: Python
License: No license data
Last GitHub push: Sep 9, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed
PR merge metrics: No merged PRs in 30d

Beginner friendly Everything

Open All

21 open issues indexed

uniform_1d: cudaEventElapsedTime raises cudaErrorNotReady as an untyped RuntimeError Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA/cuEquivariance#302 · Opened Sep 14, 2026
Trying to run boltz2 on AMD GPUS Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

NVIDIA/cuEquivariance#294 · Opened Jul 22, 2026
Request for source code access to cuequivariance-ops-cu12 and cuequivariance-ops-torch-cu12 Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

NVIDIA/cuEquivariance#292 · Opened Jul 17, 2026
[ALERT] Access to NVIDIA's Self-Hosted Runners is Expiring Open

nvidia-runners

Difficulty 2/5 1-3 hours Newbie friendliness 35/100

NVIDIA/cuEquivariance#289 · Opened Jun 30, 2026
Spherical Harmonics Call Fails Internally For 2nd Order Derivatives Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

NVIDIA/cuEquivariance#286 · Opened Jun 4, 2026
cuequivariance_torch.Linear failed in AOTI export Open

Difficulty 3/5 1-2 days Newbie friendliness 55/100

NVIDIA/cuEquivariance#282 · 1 comment · Opened May 22, 2026
`SphericalHarmonics` naive backend does not support `torch.compile` Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

NVIDIA/cuEquivariance#265 · 1 comment · Opened Apr 8, 2026
Exception when using `FullyConnectedTensorProductConv`'s second order derivatives Open

feature request

Difficulty 4/5 3-5 days Newbie friendliness 48/100

NVIDIA/cuEquivariance#264 · 1 comment · Opened Apr 6, 2026
Out of shared memory on blackwell architectures with embedding size 128 Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA/cuEquivariance#244 · 1 comment · Opened Jan 28, 2026
`torch.compile` crashes when sizes are unbacked `SymInt`s. Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

NVIDIA/cuEquivariance#243 · Opened Jan 28, 2026
Typing issue for triangle_multiplicative_update when using cuEquivariance with openfold2 Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

NVIDIA/cuEquivariance#229 · Opened Jan 7, 2026
Conda packages for cuequivariance-ops-* Open

feature request

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA/cuEquivariance#218 · 1 comment · 1 reaction · Opened Dec 1, 2025
cuequivariance-ops-cu11 for aarch64 Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

NVIDIA/cuEquivariance#215 · Opened Nov 22, 2025
Bug: cuEquivariance-jax not support the B200 / B300 | cuda12 Open

Difficulty 4/5 3-5 days Newbie friendliness 42/100

NVIDIA/cuEquivariance#209 · 4 comments · Opened Nov 10, 2025
Finer granularity of control over triangular multiplicative update Open

feature request

Difficulty 5/5 Over a week Newbie friendliness 30/100

NVIDIA/cuEquivariance#205 · Opened Oct 23, 2025
Attention fwd/bwd issues with all zero masks Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

NVIDIA/cuEquivariance#204 · 2 comments · Opened Oct 23, 2025
ubuntu24.04-wsl，pynvml.NVMLError_NotSupported: Not Supported Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

NVIDIA/cuEquivariance#203 · Opened Oct 23, 2025
Import error for cuequivariance_ops_torch even in a brand new conda env Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA/cuEquivariance#193 · 4 comments · Opened Oct 8, 2025
`AttentionPairBiasMask.backward()` fails with non-contiguous gradient tensors despite contiguous forward inputs Open

Difficulty 2/5 1-3 hours Newbie friendliness 48/100

NVIDIA/cuEquivariance#177 · 2 comments · Opened Sep 18, 2025
pynvml is not supported on GeForce Open

bug feature request

Difficulty 4/5 3-5 days Newbie friendliness 38/100

NVIDIA/cuEquivariance#165 · 5 comments · Opened Sep 1, 2025
cuet.SegmentedPolynomial does not support fp16/bf16 for math_dtype? Open

feature request

Difficulty 4/5 3-5 days Newbie friendliness 35/100

NVIDIA/cuEquivariance#143 · 4 comments · Opened Jul 28, 2025

NVIDIA/cuEquivariance

Get new issues in your inbox