-
Notifications
You must be signed in to change notification settings - Fork 231
Pull requests: NVIDIA/cuvs
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Generate cubins from New feature or request
non-breaking
Introduces a non-breaking change
cutile at build time and embed them in C++ DSO
feature request
#2548
opened Sep 3, 2026 by
divyegala
Contributor
Loading…
Keep the Fern API reference in sync with the sources
doc
Improvements or additions to documentation
non-breaking
Introduces a non-breaking change
#2547
opened Sep 3, 2026 by
imotov
Contributor
Loading…
Move Java example projects under examples/java/
improvement
Improves an existing functionality
non-breaking
Introduces a non-breaking change
#2543
opened Sep 2, 2026 by
imotov
Contributor
Loading…
Enable ARM (aarch64) support for cuvs-java and cuvs-lucene
improvement
Improves an existing functionality
non-breaking
Introduces a non-breaking change
CAGRA-ACE: Support Direct HNSW Index Creation During Build
improvement
Improves an existing functionality
non-breaking
Introduces a non-breaking change
#2539
opened Sep 1, 2026 by
julianmi
Contributor
Loading…
Optimize transfer/compute overlap in out-of-core KMeans
improvement
Improves an existing functionality
non-breaking
Introduces a non-breaking change
#2538
opened Sep 1, 2026 by
viclafargue
Contributor
Loading…
Keep the polynomial and RBF Gram epilogues in the input precision
#2536
opened Aug 31, 2026 by
maxwbuckley
Contributor
Loading…
Keep the sparse Jensen-Shannon post-processing in the input precision
#2535
opened Aug 31, 2026 by
maxwbuckley
Contributor
Loading…
Remove trailing new lines from GTest names
bug
Something isn't working
non-breaking
Introduces a non-breaking change
#2534
opened Aug 31, 2026 by
robertmaynard
Contributor
Loading…
Bound CAGRA hash table sizing loops so they cannot spin forever
#2533
opened Aug 31, 2026 by
shaunakkapur
Contributor
Loading…
[Experimental] Implement dataset as a single type+specs
breaking
Introduces a breaking change
feature request
New feature or request
Compute the expanded L2 distance in fp32 instead of fp64
C++
improvement
Improves an existing functionality
non-breaking
Introduces a non-breaking change
#2531
opened Aug 31, 2026 by
maxwbuckley
Contributor
Loading…
Reuse raft::mdarray/mdspan directly in dense dataset storage
#2530
opened Aug 30, 2026 by
HowardHuang1
Contributor
Loading…
Fall back to cudaMalloc on devices without stream-ordered allocation
#2528
opened Aug 29, 2026 by
drzraf
Loading…
Use JIT-LTO kernel handles through an API the toolkit actually provides
#2527
opened Aug 29, 2026 by
drzraf
Loading…
[ANN_BENCH] Remove Stream Pool Creation from Individual Wrappers
improvement
Improves an existing functionality
non-breaking
Introduces a non-breaking change
#2526
opened Aug 29, 2026 by
tarang-jain
Contributor
Loading…
Migrate stream APIs from rmm::cuda_stream_view to cuda::stream_ref
#2521
opened Aug 28, 2026 by
bdice
Contributor
Loading…
Use raft::launch_kernel in cagra search JIT kernels
Enhancement
improvement
Improves an existing functionality
non-breaking
Introduces a non-breaking change
Lucene: define and validate CAGRA parameter boundaries
improvement
Improves an existing functionality
non-breaking
Introduces a non-breaking change
#2516
opened Aug 27, 2026 by
shaunakkapur
Contributor
Loading…
Fix stream ordering for brute-force L2 sqrt post-processing
bug
Something isn't working
non-breaking
Introduces a non-breaking change
#2509
opened Aug 26, 2026 by
dantegd
Contributor
Loading…
Previous Next
ProTip!
Find all pull requests that aren't related to any open issues with -linked:issue.