Search results

[jax_experimental] Incorrect gradients in jax_kernel(..., enable_backward=True): launch_dims auto-inference uses full array shape and the explicit workaround is blocked

bugSomething isn't workingSomething isn't working

Status: Open.

#1380 In NVIDIA/warp;

· kshy0519 opened on Apr 21, 2026

[REQ] Add adjoint for out-of-place wp.tile_upper_solve()

autodiff

Status: Open.

Enhancement

#1379 In NVIDIA/warp;

· ns5678 opened on Apr 21, 2026

[REQ] Add adjoint for out-of-place wp.tile_lower_solve()

autodiff

Status: Open.

Enhancement

#1378 In NVIDIA/warp;

· ns5678 opened on Apr 21, 2026

[REQ] Add adjoint for out-of-place wp.tile_cholesky_solve()

autodiff

Status: Open.

Enhancement

#1377 In NVIDIA/warp;

· ns5678 opened on Apr 21, 2026

Potential bug in wp.min(a,b) when either a or b is Nan

bugSomething isn't workingSomething isn't working

Status: Open.

Bug

#1376 In NVIDIA/warp;

· gyeomannvidia opened on Apr 20, 2026

·  

1.14.0

radix_sort_pairs can cause CUDA errors when used within graph capture

bugSomething isn't workingSomething isn't working

Status: Open.

Bug

#1373 In NVIDIA/warp;

· camevor opened on Apr 17, 2026

Built-in mesh watertightness

feature requestRequest for something to be addedRequest for something to be added

Status: Open.

#1371 In NVIDIA/warp;

· xiaozxiong opened on Apr 16, 2026

Promote warp.jax_experimental API to warp.jax

interopInteroperability of Warp with other librariesInteroperability of Warp with other libraries

Status: Open.

Task

#1370 In NVIDIA/warp;

· shi-eric opened on Apr 16, 2026

·  

1.14.0

Optimize tiled CUDA launches by reading blockIdx.x directly

tile

Status: Open.

Task

#1361 In NVIDIA/warp;

· adenzler-nvidia opened on Apr 13, 2026

Split ASV CI job into three parallel jobs by variability tier

Status: Open.

#1359 In NVIDIA/warp;

· c0d1f1ed opened on Apr 12, 2026

·  

Backlog

Enable LLVM auto-vectorization for CPU kernels

feature requestRequest for something to be addedRequest for something to be added

Status: Open.

Enhancement

#1358 In NVIDIA/warp;

· c0d1f1ed opened on Apr 12, 2026

·  

1.15.0

Reduce Python-side overhead for kernel launches using METH_FASTCALL

core

Status: Open.

Enhancement

#1357 In NVIDIA/warp;

· c0d1f1ed opened on Apr 10, 2026

·  

1.14.0