All issues

Search results

GPU tropical path: keep tensors device-resident + device-side permute (replicate Julia's CuArray data flow)

agent:ready

Status: Open.

#52 In tensor4all/omeinsum-rs;

· isPANN opened on Jun 11, 2026

CPU contractions do not scale across threads (t1 ≈ t8)

agent:ready

Status: Open.

#51 In tensor4all/omeinsum-rs;

· isPANN opened on Jun 11, 2026

Cuda backend silently returns standard (+, ×) results for tropical algebras — dispatch ignores the algebra parameter (follow-up to #6)

agent:ready

Status: Open.

#48 In tensor4all/omeinsum-rs;

· isPANN opened on May 20, 2026

Profile remaining slow binary benchmark cases

agent:ready

Status: Open.

#41 In tensor4all/omeinsum-rs;

· GiggleLiu opened on Apr 18, 2026

update dependencies

Status: Open.

#28 In tensor4all/omeinsum-rs;

· GiggleLiu opened on Feb 5, 2026

Add alpha/beta scaling factors for einsum operations

Status: Open.

#23 In tensor4all/omeinsum-rs;

· GiggleLiu opened on Jan 28, 2026

RFC: Distributed Computing Support for Large-Scale Tensor Network Contraction

Status: Open.

#17 In tensor4all/omeinsum-rs;

· GiggleLiu opened on Jan 27, 2026