NVIDIA/raft

View on GitHub

RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.

Stars: 1k
Forks: 251
Open beginner issues: 0
Indexed issues: 427
Avg merge: 1d 8h
Merged PRs (30d): 31

Dominant language: Cuda
License: Apache-2.0
Last GitHub push: Sep 17, 2026
Latest indexed: Sep 19, 2026
Contributing guide: No contributing guide
Code of conduct: No code of conduct
Beginner labels: good first issue

Beginner friendly Everything

Open All

427 open issues indexed

[DOC] RAPIDS -> nvidia docs migration TODOs Open

doc

Difficulty 4/5 3-5 days Newbie friendliness 55/100

NVIDIA/raft#3150 · 1 comment · Opened Sep 11, 2026
[CI] check-nightly-ci failing Open

bug

Difficulty 1/5 Under an hour Newbie friendliness 20/100

NVIDIA/raft#3037 · Opened May 26, 2026
[TRACKER] CI Test Failures Needing attention Open

bug

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

NVIDIA/raft#3035 · Opened May 21, 2026
[FEA] Lazy initialization of a non-empty thread pool resource Open

feature request

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

NVIDIA/raft#3022 · Opened May 18, 2026
[CI] Lanczos SM tests fail cuda 13.2 Open

bug

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

NVIDIA/raft#3021 · 6 comments · Opened May 17, 2026
[FEA] Add an option argument to raft::linalg::gemm for specifying compute type Open

feature request

Difficulty 4/5 3-5 days Newbie friendliness 45/100

NVIDIA/raft#2988 · Opened Mar 20, 2026
[ENH] Consider switching to cuda::managed_memory_pool / cuda::pinned_memory_pool as the default managed/pinned resources Open

feature request

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

NVIDIA/raft#2976 · 3 comments · Opened Mar 6, 2026
The purpose of using cusolver*Host Open

question

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

NVIDIA/raft#2965 · Opened Feb 24, 2026
Update lanczos output eigenpairs to use `int64` types Open

NVIDIA/raft#2925 · 1 assignee · Opened Jan 16, 2026
Improve `device_resources_snmg` RMM pool robustness Open

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

NVIDIA/raft#2922 · Opened Jan 15, 2026
COO input for lanczos solver should be sorted Open

documentation

NVIDIA/raft#2874 · 1 assignee · Opened Nov 19, 2025
Is `CUDA_ENFORCE_LOWER` still needed? Open

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

NVIDIA/raft#2856 · Opened Oct 31, 2025
Update outdated workaround for `cusparseSpMV` Open

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

NVIDIA/raft#2854 · 1 comment · Opened Oct 30, 2025
[FEA] Migrate hash strategy to use the new cuco::static_map Open

feature request

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

NVIDIA/raft#2846 · Opened Oct 22, 2025
Releasing GIL during CUDA stream synchronization Open

feature request

Difficulty 5/5 Over a week Newbie friendliness 35/100

NVIDIA/raft#2841 · Opened Oct 17, 2025
[FEA] Remove UCX code in favor UCXX implementation Open

feature request

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

NVIDIA/raft#2792 · 1 comment · Opened Aug 26, 2025
[FEA] New KVP Types for map_kernel Open

feature request

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

NVIDIA/raft#2782 · Opened Aug 11, 2025
[BUG] Eigen solver fails to converge to a solution Open

bug

NVIDIA/raft#2758 · 1 comment · 1 assignee · Opened Jul 23, 2025
[FEA] Add dlpack utilities from cuvs Open

feature request

Difficulty 4/5 3-5 days Newbie friendliness 32/100

NVIDIA/raft#2755 · Opened Jul 22, 2025
[FEA] Lanczos Solver SM algorithm Open

cpp feature request

Difficulty 5/5 Over a week Newbie friendliness 35/100

NVIDIA/raft#2705 · Opened Jun 16, 2025
[FEA] Complete Deprecation of raft::sparse::COO from public APIs Open

feature request

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

NVIDIA/raft#2685 · 4 comments · Opened May 22, 2025
[BUG] illegal memory access due to negative offset in map_kernel on large array (1.8 billion elements) Open

bug

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

NVIDIA/raft#2683 · Opened May 22, 2025
[BUG] illegal memory access error in stridedSummationKernel when running it on a large dataset (130 million rows, 27 columns) Open

bug

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

NVIDIA/raft#2646 · Opened Apr 30, 2025
Wheel sizes and inspection after merge of https://github.com/rapidsai/raft/pull/2629 Open

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

NVIDIA/raft#2637 · 2 comments · Opened Apr 16, 2025
[BUG] error: namespace "raft::detail" has no member "format" when `RAFT_LOG_ACTIVE_LEVEL` = `RAPIDS_LOGGER_LOG_LEVEL_TRACE` Open

bug

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

NVIDIA/raft#2622 · 1 reaction · Opened Mar 29, 2025
[BUG]Run error for nvidia v100 on branch-24.04 Open

bug

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

NVIDIA/raft#2599 · 1 comment · Opened Mar 10, 2025
[BUG] called a variable which not passed to fucntion when performing RAFT_LOG_DEBUG macro Open

bug

Difficulty 1/5 Under an hour Newbie friendliness 45/100

NVIDIA/raft#2593 · Opened Feb 25, 2025
[FEA] Provide const_view_type for csr_matrix Open

feature request

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

NVIDIA/raft#2585 · Opened Feb 14, 2025
[FEA] Add new sparse matrix API to Python Open

feature request

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

NVIDIA/raft#2504 · 1 comment · Opened Nov 21, 2024
[FEA] Support the cached and reusable `sparsity` mechanism Open

feature request

NVIDIA/raft#2492 · 1 assignee · Opened Nov 14, 2024
[FEA] Find a way to generate and validate test data without hardcoding for Lanczos solver Open

feature request

Difficulty 5/5 Over a week Newbie friendliness 20/100

NVIDIA/raft#2485 · Opened Nov 5, 2024
[FEA] Raft primitives reduce_by_key Open

feature request

Difficulty 4/5 3-5 days Newbie friendliness 30/100

NVIDIA/raft#2477 · Opened Oct 30, 2024
Raft Cpp Test Error Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

NVIDIA/raft#2476 · Opened Oct 24, 2024
[FEA] Remove dependency of `build_comms_nccl_only` on UCP Open

feature request

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

NVIDIA/raft#2465 · Opened Oct 3, 2024
[FEA] rename ASSERT in error.hpp to RAFT_ASSERT Open

feature request

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

NVIDIA/raft#2462 · 3 comments · Opened Sep 30, 2024
[FEA] Improvements of NCCL clique helper functions Open

feature request

Difficulty 4/5 3-5 days Newbie friendliness 32/100

NVIDIA/raft#2459 · Opened Sep 27, 2024
[FEA] Investigate the performance issue of `cublasLtMatrixTransform ` and switch the `transpose` to it ASAP. Open

feature request

NVIDIA/raft#2436 · 1 assignee · Opened Sep 17, 2024
[FEA] Expose algorithm selection in `matrix.select_k` python API Open

feature request

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

NVIDIA/raft#2427 · Opened Sep 4, 2024
(Flaky?) C++ test `CLUSTER_TEST` Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 30/100

NVIDIA/raft#2418 · Opened Aug 23, 2024
[QST] How to use RAFT installed by mamba? Open

question

Difficulty 3/5 1-2 days Newbie friendliness 25/100

NVIDIA/raft#2412 · 1 comment · Opened Aug 13, 2024
[BUG]An illegal instruction caused by `masked_matmul` on v100 when the dim isn't multiple of 32. Open

bug

NVIDIA/raft#2411 · 1 assignee · Opened Aug 9, 2024
[BUG] next_float and next_double return values in range [0,1] Open

bug

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

NVIDIA/raft#2400 · 1 comment · Opened Jul 30, 2024
[FEA] conan support please Open

feature request

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

NVIDIA/raft#2396 · 1 comment · Opened Jul 25, 2024
Updating RAFT ANN benchmark environment/containers to newer CUDA Open

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

NVIDIA/raft#2383 · Opened Jul 17, 2024
[FEA] Support setting the bits in a bitmap filter via a user-defined CUDA functor. Open

feature request

NVIDIA/raft#2374 · 1 assignee · Opened Jul 2, 2024
[FEA] Consolidate SUM reductions across raft Open

feature request

NVIDIA/raft#2366 · 1 assignee · Opened Jun 25, 2024
[BUG] Enable logging macros in downstream projects Open

bug

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

NVIDIA/raft#2357 · Opened Jun 10, 2024
[BUG] Error when running raft_ann_bench Open

bug

Difficulty 2/5 1-3 hours Newbie friendliness 38/100

NVIDIA/raft#2356 · Opened Jun 8, 2024
[FEA]creating the masked_matmul function to front the SDDMM+custom kernel. Open

feature request

NVIDIA/raft#2336 · 1 assignee · Opened May 23, 2024
[FEA] use of COO to balance the `faster_dot kernel` across the blocks Open

feature request

NVIDIA/raft#2335 · 1 assignee · Opened May 23, 2024

NVIDIA/raft

Get new issues in your inbox