NVIDIA/warp

View on GitHub

A Python framework for GPU-accelerated simulation, robotics, and machine learning.

Stars: 7.1k
Forks: 624
Open beginner issues: 0
Indexed issues: 282
Avg merge: 3d 17h
Merged PRs (30d): 5

Dominant language: Python
License: Apache-2.0
Last GitHub push: Sep 19, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: Code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

282 open issues indexed

Improve the error when NumPy attempts to convert a CUDA wp.array Open

docs

NVIDIA/warp#1972 · 1 assignee · Opened Sep 17, 2026
Reduce redundant radix sorting in CUDA BSR transpose Open

warp.sparse

NVIDIA/warp#1971 · 1 assignee · Opened Sep 17, 2026
Support first-sample FEM interpolation with row compression Open

warp.fem

NVIDIA/warp#1970 · 1 assignee · Opened Sep 17, 2026
@wp.kernel(launch_bounds=...) fails under wp.config.llvm_cuda Open

Difficulty 2/5 1-3 hours Newbie friendliness 88/100

NVIDIA/warp#1968 · Opened Sep 17, 2026
Remove deprecated `*_tiled` tile query aliases Open

spatial acceleration

NVIDIA/warp#1967 · 1 assignee · Opened Sep 17, 2026
tile_matmul(): use the cuBLASDx register-accumulator API with suggested shared layouts for Tensor Core precisions Open

tile

NVIDIA/warp#1965 · 3 comments · 1 assignee · Opened Sep 16, 2026
graph_coloring_balance can loop forever on an equal-ratio coloring cycle Open

bug

NVIDIA/warp#1964 · 1 comment · 1 assignee · Opened Sep 15, 2026
HIP: stream capture error paths diverge from CUDA; hipErrorStreamCaptureUnjoined leaves the stream unusable Open

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

NVIDIA/warp#1957 · 2 comments · Opened Sep 14, 2026
Enable CPU code vectorization Open

feature request runtime

NVIDIA/warp#1954 · 1 comment · 1 assignee · Opened Sep 13, 2026
Blue noise sampling on surfaces Open

feature request warp.geometry

NVIDIA/warp#1951 · 1 assignee · Opened Sep 11, 2026
Oriented bounding box Open

feature request warp.geometry

NVIDIA/warp#1950 · 1 assignee · Opened Sep 11, 2026
Evaluate augmented-assignment targets before the RHS Open

bug

NVIDIA/warp#1947 · 1 comment · 1 assignee · Opened Sep 11, 2026
Preserve ordering between deterministic atomics and stores Open

bug

NVIDIA/warp#1946 · 1 assignee · Opened Sep 11, 2026
Avoid null writes in deterministic component adjoints Open

bug

NVIDIA/warp#1945 · 1 assignee · Opened Sep 11, 2026
[BUG] CUDA 13.1+ can make later Warp kernels fail on sm_120 Open

bug

NVIDIA/warp#1944 · 1 comment · 1 assignee · Opened Sep 11, 2026
Investigate remote cache support for kernel and LTO artifacts Open

feature request

NVIDIA/warp#1942 · 1 reaction · 1 assignee · Opened Sep 10, 2026
Fix missing block synchronization in tile Cholesky and matmul adjoints Open

bug

NVIDIA/warp#1941 · 1 comment · 1 assignee · Opened Sep 10, 2026
Fuse closest-cell traversal over Warp BVH Open

feature request spatial acceleration

NVIDIA/warp#1924 · 1 comment · 1 assignee · Opened Sep 5, 2026
Add reusable fixed-topology FEM assembly Open

feature request warp.fem

NVIDIA/warp#1923 · 1 comment · 1 assignee · Opened Sep 5, 2026
wp.map() raises UnboundLocalError when out is a list for a single-output function Open

Difficulty 2/5 1-3 hours Newbie friendliness 88/100

NVIDIA/warp#1913 · Opened Sep 4, 2026
Native `tile_gemm` function for pipelined GEMM over global arrays Open

feature request

NVIDIA/warp#1907 · 2 comments · 1 assignee · Opened Sep 4, 2026
Introduce an explicit tile storage model for Warp tiles. Open

feature request

NVIDIA/warp#1901 · 1 assignee · Opened Sep 3, 2026
Build from source fails to find MSVC on Build Tools-only installations (vswhere omits Build Tools without -products *) Open

bug

NVIDIA/warp#1892 · 1 comment · 1 assignee · Opened Sep 2, 2026
Add Block Jacobi Preconditioner for Warp CG Solver Open

feature request warp.sparse

NVIDIA/warp#1891 · 1 assignee · Opened Sep 2, 2026
warp.optim: Adam and SGD validate gradient shape with bare assert, so -O skips the check Open

Difficulty 2/5 1-3 hours Newbie friendliness 82/100

NVIDIA/warp#1880 · Opened Aug 30, 2026
Reduce `wp.launch()` Host Overhead Open

runtime

NVIDIA/warp#1862 · 1 comment · 2 assignees · Opened Aug 26, 2026
Clarify memory sharing and synchronization in interoperability APIs Open

docs

NVIDIA/warp#1859 · 1 assignee · Opened Aug 26, 2026
Document array atomic behavior and deterministic execution Open

docs

NVIDIA/warp#1858 · 1 assignee · Opened Aug 26, 2026
Document small-matrix decomposition and inversion behavior Open

docs

NVIDIA/warp#1857 · 1 assignee · Opened Aug 26, 2026
Document quaternion conventions and preconditions Open

docs

NVIDIA/warp#1856 · 1 assignee · Opened Aug 26, 2026
Document and link generic notation in built-in signatures Open

docs

NVIDIA/warp#1855 · 1 assignee · Opened Aug 26, 2026
Document wp.Int, wp.Float, and wp.Scalar Open

docs

NVIDIA/warp#1854 · 1 assignee · Opened Aug 26, 2026
Add unit-quaternion support to forward-mode jets Open

autodiff feature request

NVIDIA/warp#1847 · 1 assignee · Opened Aug 25, 2026
BVH volume (AABB/sphere) traversal spends 20-30% in stack traffic and node reloads Open

spatial acceleration

NVIDIA/warp#1846 · 1 assignee · Opened Aug 24, 2026
BVH ray/capsule queries carry a traversal stack and 32KB shared memory they don't need Open

spatial acceleration

NVIDIA/warp#1845 · 1 assignee · Opened Aug 24, 2026
[Tracking] BVH and mesh query performance improvements Open

spatial acceleration

NVIDIA/warp#1842 · 1 assignee · Opened Aug 24, 2026
Python built-ins accept keywords from the wrong overload Open

bug

NVIDIA/warp#1841 · 1 assignee · Opened Aug 24, 2026
APIC portability onto other devices Open

feature request

NVIDIA/warp#1837 · 1 comment · 1 assignee · Opened Aug 24, 2026
[BUG] mesh_query_point_sign_normal reports exterior points as inside a tetrahedron Open

bug

NVIDIA/warp#1836 · 2 comments · 1 assignee · Opened Aug 24, 2026
Add a warp function call hook to detect when warp functions are being called. Open

feature request

NVIDIA/warp#1834 · 1 assignee · Opened Aug 24, 2026
Loaded APIC graphs cannot join a caller-owned CUDA capture, and child-graph nesting is refused for conditional nodes Open

feature request question

NVIDIA/warp#1827 · 3 comments · 1 assignee · Opened Aug 21, 2026
Establish the initial `warp.geometry` module Open

feature request

NVIDIA/warp#1826 · 1 assignee · Opened Aug 21, 2026
Deterministic mode silently corrupts scratch-dependent counter reservations Open

bug

NVIDIA/warp#1817 · 1 assignee · Opened Aug 18, 2026
second-order forward mode autodiff Open

autodiff feature request

NVIDIA/warp#1813 · 1 assignee · Opened Aug 17, 2026
CUDA: incorrect `Texture3D` samples when texture handles vary across lanes Open

bug

NVIDIA/warp#1811 · 3 comments · 1 assignee · Opened Aug 17, 2026
FEM example: Trimesh2D side topology goes stale after in-place edge flips Open

bug warp.fem

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

NVIDIA/warp#1804 · Opened Aug 14, 2026
Graph-owned host regions for APIC CUDA graphs Open

apic feature request

NVIDIA/warp#1801 · 1 assignee · Opened Aug 14, 2026
Loaded APIC graph wraps scalar wp.tid() for dim=(2**31, 2) Open

apic bug

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

NVIDIA/warp#1800 · Opened Aug 14, 2026
ModuleHasher: kernel hash unstable across multi-device build (dense overload graphs) Open

bug runtime

NVIDIA/warp#1798 · 2 comments · 1 assignee · Opened Aug 14, 2026
wp.grad() misses custom @wp.func_grad implementations Open

bug

NVIDIA/warp#1794 · 1 assignee · Opened Aug 14, 2026

NVIDIA/warp

Get new issues in your inbox