v25.06.00

🚨 Breaking Changes

• [Java] Adding support for prefiltering in CAGRA (#870) @punAhuja

🐛 Bug Fixes

• Fix failing CAGRA merge google tests for 25.06 (#974) @rhdong
• Fix for recent NCCL resource update (#968) @viclafargue
• Revert "Fix kmeans::predict argument order (#915)" (#951) @divyegala
• IVF-PQ tests: fix segfault when accessing empty lists (#933) @achirkin
• batch_load_iterator shall copy data located in host memory (#926) @tfeher
• Fix kmeans::predict argument order (#915) @enp1s0
• ANN_BENCH: fix the reported core count (#896) @achirkin
• CUVS_ANN_BENCH_SINGLE_EXE: find the mg library (#890) @achirkin
• [FIX] Revert negated InnerProduct distance for NN Descent (#859) @jinsolp
• IVF-PQ tests: fix segfault when accessing empty lists (#838) @achirkin
• [CuVS-Java] Automate panama bindings generation, Include IVF_PQ parameters in CAGRA index parameters and other changes (#831) @narangvivek10
• Ann-bench: fix unsafe lazy blobs (#828) @achirkin
• Fix test_brute_force conversion (#821) @lowener
• [FEA] New algos and updates/corrections to Faiss cuvs-bench (#677) @tarang-jain

📖 Documentation

• All-neighbors API docs (#944) @jinsolp
• cagra.rst: fitered -> filtered (#866) @eli-b
• [DOCS] Fix Integration Docs (Faiss) (#782) @tarang-jain

🚀 New Features

• Add IVF-PQ support in CAGRA index params in Python (#918) @lowener
• [Feat] Expose C API for CAGRA merge (#860) @rhdong
• Use NCCL wheels from PyPI for CUDA 12 builds (#827) @divyegala
• Add support for half in CAGRA+HNSW (#813) @lowener
• [CI] Enable Java test in CI workflow (#805) @rhdong
• Wrapper for all-neighbors knn graph building (#785) @jinsolp
• Add support of half dtype in IVF-FLAT (#730) @lowener
• IVF-PQ: low-precision coarse search (#715) @achirkin
• [Feat] Add support of logical merge in Cagra (#713) @rhdong

🛠️ Improvements

• Update score calculation for CAGRA-Q instance selection (#938) @enp1s0
• [FEA] Use Native Brute Force for Sparse Pairwise KNN (#927) @tarang-jain
• use 'rapids-init-pip' in wheel CI, other CI changes (#917) @jameslamb
• ANN_BENCH: Avoid repeated calls to raft::get_device_for_address in CAGRA search (#908) @achirkin
• [Java] New off-heap Dataset support for CAGRA and Bruteforce (#902) @chatman
• Finish CUDA 12.9 migration and use branch-25.06 workflows (#901) @bdice
• Update to clang 20 (#898) @bdice
• get Java artifacts from GitHub Actions artifact store (#893) @jameslamb
• Quote head_rev in conda recipes (#892) @bdice
• [Java] Exposing merge API for multiple CAGRA indices (#891) @punAhuja
• Expose ivf-flat centers to python/c (#888) @benfred
• CUDA 12.9 use updated compression flags (#887) @robertmaynard
• Expose ivf-pq centers to python/c (#881) @benfred
• Accept host inputs in python for ivf-pq build and extend (#880) @benfred
• Add tiered_index support (#879) @benfred
• Exclude librmm.so from auditwheel (#878) @bdice
• update.version.sh: remove broken reference, skip most CI on PRs that only modify update-version.sh (#875) @jameslamb
• Add support for Python 3.13 (#874) @gforsyth
• chore: lower wheel size threshold (#872) @gforsyth
• [Java] Adding support for prefiltering in CAGRA (#870) @punAhuja
• Change snmg index to use updated multi gpu resource API (#869) @jinsolp
• run shellcheck on all files, other small pre-commit updates (#865) @jameslamb
• Fix IVF PQ build metric for CAGRA (#862) @lowener
• ANN_BENCH: Expose parallel_mode parameter of FAISS CPU IVF implementation (#861) @achirkin
• Specify matplotlib version (#839) @benfred
• Use random tmp names for index files in tests (#837) @achirkin
• Download build artifacts from Github for CI (#834) @VenkateshJaya
• Add NN-Descent return_distances functionality to python/C (#833) @benfred
• Optimize hnsw::from_cagra<GPU> (#826) @achirkin
• Use vendored RAPIDS.cmake in example code. (#824) @bdice
• refactor(rattler): enable strict channel priority for builds (#823) @gforsyth
• Reduce device memory usage for CAGRA's graph optimization process (2-hop detour counting) (#822) @anaruse
• [cuvs_bench] distinguish search label from build label in data_export.py (#818) @jiangyinzuo
• Vendor RAPIDS.cmake (#816) @bdice
• Update libcuvs libraft ver to 25.06 in conda env (#808) @jinsolp
• Moving NN Descent class and struct declarations to nn_descent_gnnd.hpp (#803) @jinsolp
• Remove @rapidsai/cuvs-build-codeowners (#783) @KyleFromNVIDIA
• Moving wheel builds to specified location and uploading build artifacts to Github (#777) @VenkateshJaya
• Remove unused raft cagra header in add_nodes.cuh (#741) @jiangyinzuo
• Expose kmeans to python (#729) @benfred
• Update cuvs to properly create a NCCL::NCCL target (#720) @robertmaynard
• Optimize euclidean distance in host refine phase (#689) @anstellaire
• Moving MG functions into unified API + raft::device_resources_snmg as device resource type for MG functions (#454) @viclafargue
• Moving random ball cover (#218) @cjnolet

[NIGHTLY] v25.08.00

🔗 Links

• Development Branch
• Compare with main branch

🚨 Breaking Changes

• Remove CUDA 11 from dependencies.yaml (#962) @KyleFromNVIDIA
• stop uploading packages to downloads.rapids.ai (#940) @jameslamb

🐛 Bug Fixes

• Pin Faiss Version to Last Stable Release (#1029) @tarang-jain
• Fix verify-copyright precommit hook (#1017) @benfred
• Fix C compile error in tiered_index.h (#996) @mythrocks
• Fix for NN Descent negative or NaN distances (#994) @jinsolp
• Adding int64 search for MG CAGRA (#975) @viclafargue
• [BUG] cuvs-bench data_export fixes (#863) @tarang-jain

📖 Documentation

• [REVIEW][Java] Add instructions on how to run a single test (#1004) @ldematte

🚀 New Features

• Add C-API method to fetch cuVS version from Java. (#935) @mythrocks

🛠️ Improvements

• Update nightly CI check to allow 30 days of failing nightly build. (#1008) @cjnolet
• [REVIEW][Java] Refactor CagraBuildAndSearchIT to explicitly express different execution modes (#1006) @ldematte
• [REVIEW][Java] Add jextract artifacts to gitignore (#1005) @ldematte
• Remove pytest pin (#998) @vyasr
• [java] Utility function for cudaMemcpy (#983) @mythrocks
• [java] Copy pom.xml to target/ on build (#981) @mythrocks
• Remove CUDA 11 from dependencies.yaml (#962) @KyleFromNVIDIA
• refactor(rattler): remove cuda11 options and general cleanup (#961) @gforsyth
• Remove CUDA 11 devcontainers and update CI scripts (#960) @bdice
• stop uploading packages to downloads.rapids.ai (#940) @jameslamb
• Instantiate only specific RAFT reduction kernels (#925) @divyegala
• [Java] Replace the C wrapper layer with the Panama API-based Java implementation (#916) @narangvivek10
• Forward-merge branch-25.06 into branch-25.08 (#909) @gforsyth
• Forward-merge branch-25.06 into branch-25.08 (#897) @gforsyth

v25.04.00

🚨 Breaking Changes

• Use new rapids-logger library (#644) @vyasr

🐛 Bug Fixes

• Use new build_patch_only argument (#780) @KyleFromNVIDIA
• Fix resource import so generate_groundtruth works (#774) @nvrohanv
• Relax max duplicates in batched NN Descent (#770) @jinsolp
• [BUG] Fix graph index sorting in CAGRA graph build by NN Descent (#763) @enp1s0
• cuvs-bench-cpu: avoid 'mkl' dependency on aarch64 (#750) @jameslamb
• ANN_BENCH: Fix segfault in CAGRA wrapper when moving its graph (#733) @achirkin
• Fix duplicate indices in batch NN Descent (#702) @jinsolp

📖 Documentation

• Go module - Usage docs (#779) @AyodeAwe

🚀 New Features

• L2SqrtExpanded metric support for NN Descent (#790) @jinsolp
• CAGRA search: int64_t indices in the API (#769) @achirkin
• Expose IVF PQ params in CAGRA C API (#768) @divyegala
• [Feat] Expose search with bitset C API for Brute Force (#717) @rhdong
• Diskann Benchmarking Wrapper (#260) @tarang-jain

🛠️ Improvements

• Use L4 instead of V100 (#797) @bdice
• Rust publishing updates (#789) @AyodeAwe
• Removing print statement (#786) @jinsolp
• Support passing a dataset to FromCagra (#767) @ajit283
• add function to build specific gpu-arches [REVIEW] (#766) @nvrohanv
• feat(libcuvs): port libcuvs to rattler-build (#751) @gforsyth
• Consolidate Conda environment for Rust (#745) @KyleFromNVIDIA
• rename 'sklearn' conda dependency to 'scikit-learn' (#743) @jameslamb
• Reduce memory consumption of scalar quantizer (#736) @mfoerste4
• Use conda-build instead of conda-mambabuild (#723) @bdice
• Expand NVTX annotations in CAGRA build and HNSW (#711) @achirkin
• Replace cub::TransformInputIterator with thrust::transform_iterator (#707) @miscco
• Consolidate more Conda solves in CI (#701) @KyleFromNVIDIA
• Require CMake 3.30.4 (#691) @robertmaynard
• Create Conda CI test env in one step (#684) @KyleFromNVIDIA
• Use shared-workflows branch-25.04 (#669) @bdice
• Add shellcheck to pre-commit and fix warnings (#662) @gforsyth
• Add build_type input field for test.yaml (#661) @gforsyth
• Use rapids-pip-retry in CI jobs that might need retries (#648) @gforsyth
• Use new rapids-logger library (#644) @vyasr
• disallow fallback to Make in Python builds (#636) @jameslamb
• Add verify-codeowners hook (#633) @KyleFromNVIDIA
• Forward-merge branch-25.02 into branch-25.04 (#632) @bdice
• Account for RAFT sparse types updates (#629) @divyegala
• Migrate to NVKS for amd64 CI runners (#627) @bdice

[NIGHTLY] v25.06.00

🔗 Links

• Development Branch
• Compare with main branch

🚨 Breaking Changes

• [Java] Adding support for prefiltering in CAGRA (#870) @punAhuja

🐛 Bug Fixes

• Update rust/cuvs-sys (#990) @benfred
• Fix rust publishing step (#984) @benfred
• Fix failing CAGRA merge google tests for 25.06 (#974) @rhdong
• Fix for recent NCCL resource update (#968) @viclafargue
• Revert "Fix kmeans::predict argument order (#915)" (#951) @divyegala
• IVF-PQ tests: fix segfault when accessing empty lists (#933) @achirkin
• batch_load_iterator shall copy data located in host memory (#926) @tfeher
• Fix kmeans::predict argument order (#915) @enp1s0
• ANN_BENCH: fix the reported core count (#896) @achirkin
• CUVS_ANN_BENCH_SINGLE_EXE: find the mg library (#890) @achirkin
• [FIX] Revert negated InnerProduct distance for NN Descent (#859) @jinsolp
• IVF-PQ tests: fix segfault when accessing empty lists (#838) @achirkin
• [CuVS-Java] Automate panama bindings generation, Include IVF_PQ parameters in CAGRA index parameters and other changes (#831) @narangvivek10
• Ann-bench: fix unsafe lazy blobs (#828) @achirkin
• Fix test_brute_force conversion (#821) @lowener
• [FEA] New algos and updates/corrections to Faiss cuvs-bench (#677) @tarang-jain

📖 Documentation

• All-neighbors API docs (#944) @jinsolp
• cagra.rst: fitered -> filtered (#866) @eli-b
• [DOCS] Fix Integration Docs (Faiss) (#782) @tarang-jain

🚀 New Features

• Add IVF-PQ support in CAGRA index params in Python (#918) @lowener
• [Feat] Expose C API for CAGRA merge (#860) @rhdong
• Use NCCL wheels from PyPI for CUDA 12 builds (#827) @divyegala
• Add support for half in CAGRA+HNSW (#813) @lowener
• [CI] Enable Java test in CI workflow (#805) @rhdong
• Wrapper for all-neighbors knn graph building (#785) @jinsolp
• Add support of half dtype in IVF-FLAT (#730) @lowener
• IVF-PQ: low-precision coarse search (#715) @achirkin
• [Feat] Add support of logical merge in Cagra (#713) @rhdong

🛠️ Improvements

• Update score calculation for CAGRA-Q instance selection (#938) @enp1s0
• [FEA] Use Native Brute Force for Sparse Pairwise KNN (#927) @tarang-jain
• use 'rapids-init-pip' in wheel CI, other CI changes (#917) @jameslamb
• ANN_BENCH: Avoid repeated calls to raft::get_device_for_address in CAGRA search (#908) @achirkin
• [Java] New off-heap Dataset support for CAGRA and Bruteforce (#902) @chatman
• Finish CUDA 12.9 migration and use branch-25.06 workflows (#901) @bdice
• Update to clang 20 (#898) @bdice
• get Java artifacts from GitHub Actions artifact store (#893) @jameslamb
• Quote head_rev in conda recipes (#892) @bdice
• [Java] Exposing merge API for multiple CAGRA indices (#891) @punAhuja
• Expose ivf-flat centers to python/c (#888) @benfred
• CUDA 12.9 use updated compression flags (#887) @robertmaynard
• Expose ivf-pq centers to python/c (#881) @benfred
• Accept host inputs in python for ivf-pq build and extend (#880) @benfred
• Add tiered_index support (#879) @benfred
• Exclude librmm.so from auditwheel (#878) @bdice
• update.version.sh: remove broken reference, skip most CI on PRs that only modify update-version.sh (#875) @jameslamb
• Add support for Python 3.13 (#874) @gforsyth
• chore: lower wheel size threshold (#872) @gforsyth
• [Java] Adding support for prefiltering in CAGRA (#870) @punAhuja
• Change snmg index to use updated multi gpu resource API (#869) @jinsolp
• run shellcheck on all files, other small pre-commit updates (#865) @jameslamb
• Fix IVF PQ build metric for CAGRA (#862) @lowener
• ANN_BENCH: Expose parallel_mode parameter of FAISS CPU IVF implementation (#861) @achirkin
• Specify matplotlib version (#839) @benfred
• Use random tmp names for index files in tests (#837) @achirkin
• Download build artifacts from Github for CI (#834) @VenkateshJaya
• Add NN-Descent return_distances functionality to python/C (#833) @benfred
• Optimize hnsw::from_cagra<GPU> (#826) @achirkin
• Use vendored RAPIDS.cmake in example code. (#824) @bdice
• refactor(rattler): enable strict channel priority for builds (#823) @gforsyth
• Reduce device memory usage for CAGRA's graph optimization process (2-hop detour counting) (#822) @anaruse
• [cuvs_bench] distinguish search label from build label in data_export.py (#818) @jiangyinzuo
• Vendor RAPIDS.cmake (#816) @bdice
• Update libcuvs libraft ver to 25.06 in conda env (#808) @jinsolp
• Moving NN Descent class and struct declarations to nn_descent_gnnd.hpp (#803) @jinsolp
• Remove @rapidsai/cuvs-build-codeowners (#783) @KyleFromNVIDIA
• Moving wheel builds to specified location and uploading build artifacts to Github (#777) @VenkateshJaya
• Remove unused raft cagra header in add_nodes.cuh (#741) @jiangyinzuo
• Expose kmeans to python (#729) @benfred
• Update cuvs to properly create a NCCL::NCCL target (#720) @robertmaynard
• Optimize euclidean distance in host refine phase (#689) @anstellaire
• Moving MG functions into unified API + raft::device_resources_snmg as device resource type for MG functions (#454) @viclafargue
• Moving random ball cover (#218) @cjnolet

v25.02.01

🚨 Breaking Changes

• [Fix] Various fixes for 25.02.01 point release (#695) @rhdong
• update pip devcontainers to UCX 1.18, small update-version.sh fixes (#604) @jameslamb
• Improve the performance of CAGRA new vector addition with the default params (#569) @enp1s0
• Update for raft logger changes (#540) @vyasr

🐛 Bug Fixes

• [Fix] Various fixes for 25.02.01 point release (#695) @rhdong
• Fix ann-bench dataset blob integer overflow leading to incorrect data copy beyond 4B elems (#671) @achirkin
• Fix ann-bench deadlocking on HNSW destruction due to task locks (#667) @achirkin
• cuvs-bench Fixes (#654) @tarang-jain
• Fix std::lock_guard use for gcc 14 support (#639) @enp1s0
• Fix indexing bug when using parallelism to build CPU hierarchy in HNSW (#620) @divyegala
• add runtime dependency on libcuvs in cuvs wheels (#615) @jameslamb
• Temporarily skip CUDA 11 wheel CI (#599) @bdice
• [Fix] l2_exp random fail in half-float32 mixed precision on self-neighboring (#596) @rhdong
• Add CAGRA InnerProduct test and fix a bug (#595) @enp1s0
• fix cuvs_bench.run --groups options (#592) @jiangyinzuo
• Fix cagra_hnsw serialization when dataset is not part of index (#591) @tfeher
• fix create_pointset for throughput mode (#589) @jiangyinzuo
• Fix the use of constexpr in the dynamic batching header (#582) @achirkin
• Reduce the recall threshold for IVF-PQ low-precision LUT inner product tests (#573) @achirkin
• Small fixes to docs and pairwise distances (#570) @cjnolet
• [BUG] Fix CAGRA graph optimization bug (#565) @enp1s0
• Fix broken link to python doc (#564) @lowener
• Fix cagra::extend error message (#532) @enp1s0
• Fix Grace-specific issues in CAGRA (#527) @achirkin

📖 Documentation

• add docs for nn_descent (#668) @Intron7
• Fixing small typo in cuvs bench docs (#586) @cjnolet
• Fix typos in README (#543) @nvanbenschoten
• Use nvidia-sphinx-theme for docs (#528) @benfred

🚀 New Features

• Add deep-100M to datasets.yaml for cuvs-bench (#670) @tarang-jain
• Expose configuration singleton as a global context for ann-bench algos (#647) @achirkin
• ANN_BENCH enhanced dataset support (#624) @achirkin
• [Feat] Add Support for Index merge in CAGRA (#618) @rhdong
• HNSW GPU hierarchy (#616) @divyegala
• CAGRA binary Hamming distance support (#610) @enp1s0
• Add cuda 12.8 support (#605) @robertmaynard
• Add support for refinement with uint32_t index type (#563) @lowener
• [Feat] Support bitset filter for Brute Force (#560) @rhdong
• Remove upper bounds on cuda-python to allow 12.6.2 and 11.8.5 (#508) @bdice

🛠️ Improvements

• [Opt] add required content to cuvs_java pom.xml (#731) @rhdong
• Add filtering to python for ivf_flat (#664) @benfred
• Expose binary quantizer to C and Python (#660) @benfred
• Add telemetry (#652) @gforsyth
• Revert docs builds to CI latest tag. (#643) @bdice
• Add float16 support in python for cagra/brute_force/ivf_pq and scalar quantizer (#637) @benfred
• Expose NN-Descent to C and Python (#635) @benfred
• Revert CUDA 12.8 shared workflow branch changes (#630) @vyasr
• cuvs-java: Rework the api to be Java 21 friendly (#628) @ChrisHegarty
• Build and test with CUDA 12.8.0 (#621) @bdice
• Add Scalar Quantization to the c and python apis (#617) @benfred
• Iteratively build graph index (#612) @anaruse
• update pip devcontainers to UCX 1.18, small update-version.sh fixes (#604) @jameslamb
• Reduce CAGRA test runtime (#602) @bdice
• Revert "Temporarily skip CUDA 11 wheel CI" (#601) @bdice
• introduce libcuvs wheels (#594) @jameslamb
• Normalize whitespace (#593) @bdice
• Rename test to tests. (#590) @bdice
• Use cuda.bindings layout. (#588) @bdice
• run_cuvs_pytests.sh uses proper test dir (#584) @robertmaynard
• expose col-major bfknn to python (#575) @benfred
• Run cuvs-bench pytests and end-to-end tests in CI (#574) @dantegd
• Expose col-major pairwise distances to python (#572) @benfred
• Improve the performance of CAGRA new vector addition with the default params (#569) @enp1s0
• Improve filtering documentation (#568) @lowener
• Use GCC 13 in CUDA 12 conda builds. (#567) @bdice
• Allow brute_force::build to work on host matrix dataset (#562) @benfred
• FAISS with cuVS enabled in cuvs-bench (#561) @tarang-jain
• Vamana build improvement and added docs (#558) @bkarsin
• Support raft's logger targets (#557) @vyasr
• Get Breathe from conda again (#554) @vyasr
• Check if nightlies have succeeded recently enough (#548) @vyasr
• Add support for float16 to the python pairwise distance api (#547) @benfred
• Additional Distances for CAGRA C and Python API (#546) @tarang-jain
• remove setup.cfg files, other packaging cleanup (#544) @jameslamb
• Fix CI for python cuvs_bench (#541) @benfred
• Update for raft logger changes (#540) @vyasr
• Change brute_force api to match ivf*/cagra (#536) @benfred
• Branch 25.02 merge 24.12 (#526) @benfred
• Update cuda-python lower bounds to 12.6.2 / 11.8.5 (#524) @bdice
• Automatic adjustment of itopk size according to filtering rate (#509) @anaruse
• prefer system install of UCX in devcontainers (#501) @jameslamb
• Adapt to rmm logger changes (#499) @vyasr
• Require approval to run CI on draft PRs (#498) @bdice
• Remove RAFT BUILD_ANN_BENCH option (#497) @bdice
• Update example code fetching rapids-cmake to use CUVS instead of RAFT (#493) @bdice
• Improve multi-CTA algorithm (#492) @anaruse
• Add filtering for CAGRA to C API (#452) @ajit283
• Initial cut for a cuVS Java API (#450) @chatman
• Add breaking change workflow trigger (#442) @AyodeAwe
• Expose extend() in C API (#276) @ajit283
• Go API - [WIP] (#212) @ajit283

v24.12.00

🚨 Breaking Changes

• HNSW CPU Hierarchy (#465) @divyegala
• Use dashes in cuvs-bench package name. (#417) @bdice

🐛 Bug Fixes

• Skip IVF-PQ packing test for lists with not enough data (#512) @achirkin
• [BUG] Fix CAGRA filter (#489) @enp1s0
• Add kIsSingleSource to PairwiseDistanceEpilogueElementwise (#485) @KyleFromNVIDIA
• Fix include errors, header, and unsafe locks in iface.hpp (#467) @achirkin
• Fix an OOB error in device-side cuvs::neighbors::refine and CAGRA kern_prune (#460) @achirkin
• Put a ceiling on cuda-python (#445) @bdice
• Enable NVTX in cuvs-cagra-search component (#439) @achirkin
• BUG: CAGRA multi-cta illegal access with bad queries (#438) @achirkin
• Fix index overflow in edge cases of CAGRA graph optimize (#435) @achirkin
• Fix correct call to brute force in generate groundtruth of cuvs-bench (#427) @dantegd
• Use Python for sccache hit rate computation. (#420) @bdice
• Add click package to cuvs-bench conda recipe (#408) @divyegala
• Fix NVTX annotations (#400) @achirkin

📖 Documentation

• [Doc] Fix CAGRA search sample code (#484) @enp1s0
• Fix broken link in README.md references (#473) @Azurethi
• Adding tech stack to docs (#448) @cjnolet
• Fix Question Retrieval notebook (#352) @lowener

🚀 New Features

• Add C++ API scalar quantization (#494) @mfoerste4
• HNSW CPU Hierarchy (#465) @divyegala
• Add serialization API to brute-force (#461) @lowener
• Add Question Retrieval notebook using Milvus (#451) @lowener
• Migrate feature diff for NN Descent from RAFT to cuVS (#421) @divyegala
• Add --no-lap-sync cmd option to ann-bench (#405) @achirkin
• Add InnerProduct and CosineExpanded metric support in NN Descent (#177) @divyegala

🛠️ Improvements

• Update cuvs to match raft's cutlass changes (#516) @vyasr
• add a README for wheels (#504) @jameslamb
• Move check_input_array from pylibraft (#474) @benfred
• use different wheel-size thresholds based on CUDA version (#469) @jameslamb
• Modify cuvs-bench to be able to generate ground truth in CPU systems (#466) @dantegd
• enforce wheel size limits, README formatting in CI (#464) @jameslamb
• Moving spectral embedding and kernel gramm APIs to cuVS (#463) @cjnolet
• Migrate sparse knn and distances code from raft (#457) @benfred
• Don't presume pointers location infers usability. (#441) @robertmaynard
• call enable_testing in root CMakeLists.txt (#437) @robertmaynard
• CAGRA tech debt: distance descriptor and workspace memory (#436) @achirkin
• Add ci run_ scripts needed for build infra (#434) @robertmaynard
• Use environment variables in cache hit rate computation. (#422) @bdice
• Use dashes in cuvs-bench package name. (#417) @bdice
• We need to enable the c_api by default (#416) @robertmaynard
• print sccache stats in builds (#413) @jameslamb
• make conda installs in CI stricter (#406) @jameslamb
• Ivf c example (#404) @abner-ma
• Prune workflows based on changed files (#392) @KyleFromNVIDIA
• [WIP] Add pinned memory resource to C API (#311) @ajit283
• Dynamic Batching (#261) @achirkin

v24.10.00

🐛 Bug Fixes

• Use 64 bit types for dataset size calculation in CAGRA graph optimizer (#380) @tfeher
• Remove EXPLICIT_INSTANTIATE_ONLY macros (#358) @achirkin
• Fix order of operations for cosine IVF Flat (#329) @lowener
• Exclude any kernel symbol that uses cutlass (#314) @benfred
• [Fix] pin raft dependent to rapidsai (#299) @rhdong
• Fix dataset dimension in IVF-PQ C wrappers (#292) @tfeher
• Fix python ivf-pq for int8/uint8 dtypes (#271) @benfred
• FP16 API for CAGRA and IVF-PQ (#264) @tfeher

📖 Documentation

• More doc updates for 24.10 (#396) @cjnolet
• fix 404 in documentation link in readme (#395) @benfred
• Improving getting started materials (#342) @cjnolet
• Fix broken examples link in README. (#326) @bdice
• Recommend miniforge for conda install (#325) @bdice

🚀 New Features

• Port remaining scripts to cuvs_bench (#368) @divyegala
• [Feat] Relative change with bitset API feature #2439 in raft (#350) @rhdong
• cuvs_bench plotting functions (#347) @dantegd
• CosineExpanded Metric for IVF-PQ (normalize inputs) (#346) @tarang-jain
• Python API for CAGRA+HNSW (#246) @divyegala
• C API for CAGRA+HNSW (#240) @divyegala
• SNMG ANN (#231) @viclafargue
• [FEA] Support for half-float mixed precise in brute-force (#225) @rhdong

🛠️ Improvements

• Remove cuvs-cagra-search from cuvs_static link (#388) @benfred
• Add a static library for cuvs (#382) @benfred
• Put the ann-bench large_workspace_resource in managed memory (#372) @achirkin
• Add multigpu kmeans fit function (#348) @benfred
• Update update-version.sh to use packaging lib (#344) @AyodeAwe
• remove NCCL pins in build and test environments (#341) @jameslamb
• Vamana/DiskANN index build (#339) @bkarsin
• Use CI workflow branch 'branch-24.10' again (#331) @jameslamb
• fix style checks on Python 3.12 (#328) @jameslamb
• Update flake8 to 7.1.1. (#327) @bdice
• Add function for calculating the mutual_reachability_graph (#323) @benfred
• Simplify libcuvs conda recipe. (#322) @bdice
• Refactor dependencies.yaml to use depends-on pattern. (#321) @bdice
• Update Python versions in cuvs_bench pyproject.toml. (#318) @bdice
• Brute force knn tile size heuristic (#316) @mfoerste4
• Euclidean distance example (#315) @abner-ma
• Migrate trustworthiness and silhouette_score stats from RAFT (#313) @benfred
• Add support for Python 3.12 (#312) @jameslamb
• Add managed option for RMM Pool memory resource to C API (#305) @ajit283
• Update rapidsai/pre-commit-hooks (#303) @KyleFromNVIDIA
• Expose search function with pre-filter for ANN (#302) @lowener
• Drop Python 3.9 support (#301) @jameslamb
• Use CUDA math wheels (#298) @KyleFromNVIDIA
• Remove NumPy <2 pin (#297) @seberg
• CAGRA - separable compilation for distance computation (#296) @achirkin
• Updating example notebooks (#294) @cjnolet
• Add RMM Pool memory resource to C API (#285) @ajit283
• Update pre-commit hooks (#283) @KyleFromNVIDIA
• Improve update-version.sh (#282) @bdice
• Use tool.scikit-build.cmake.version, set scikit-build-core minimum-version (#280) @jameslamb
• Add cuvs_bench.run python code and build (#279) @dantegd
• Add cuvs-bench to dependencies and conda environments (#275) @dantegd
• Update pip devcontainers to UCX v1.17.0 (#262) @jameslamb
• Adding example for tuning build and search params using Optuna (#257) @dpadmanabhan03
• Fixed link to build docs and corrected ivf_flat_example (#255) @mmccarty
• Merge branch-24.08 into branch-24.10 (#254) @jameslamb
• Persistent CAGRA kernel (#215) @achirkin
• [FEA] Support for Cosine distance in IVF-Flat (#179) @lowener

v24.08.00

🚨 Breaking Changes

• Allow serialization on streams (#173) @benfred

🐛 Bug Fixes

• Remove fp16 kernels that have no public entry point (#268) @tfeher
• Use raft::util::popc(...) public API (#249) @divyegala
• Enable building FAISS main statically (#241) @tarang-jain
• CAGRA bench: use device-side refinement when the data is on device (#228) @achirkin
• Rename .devcontainers for CUDA 12.5 (#224) @jakirkham
• Fix a CAGRA graph opt bug (#192) @enp1s0

📖 Documentation

• fix library name in docs ('cuvs' not 'pycuvs') (#193) @jameslamb

🚀 New Features

• Add cuvs_bench python folder, config files and constraints (#244) @dantegd
• Add MST optimization to guarantee the connectivity of CAGRA graphs (#237) @anaruse
• Moving over C++ API of CAGRA+hnswlib from RAFT (#229) @divyegala
• [FEA] expose python & C API for prefiltered brute force (#174) @rhdong
• CAGRA new vector addition (#151) @enp1s0

🛠️ Improvements

• [Opt] introduce the masked_matmul to prefiltered brute force. (#251) @rhdong
• Add more info to ANN_BENCH context (#248) @achirkin
• split up CUDA-suffixed dependencies in dependencies.yaml (#247) @jameslamb
• Fix pinning to a different RAFT tag (#235) @benfred
• Use workflow branch 24.08 again (#234) @KyleFromNVIDIA
• chore: update search_plan.cuh (#232) @eltociear
• Enable kernel & memcpy overlapping in IVF index building (#230) @abc99lr
• CAGRA: reduce argument count in select_and_run() kernel wrappers (#227) @achirkin
• Mark the rust brute force unittest as flaky (#226) @benfred
• Add python bindings for ivf-* extend functions (#220) @benfred
• Build and test with CUDA 12.5.1 (#219) @KyleFromNVIDIA
• Add col-major support for brute force knn (#217) @benfred
• Add CUDA_STATIC_MATH_LIBRARIES (#216) @KyleFromNVIDIA
• skip CMake 3.30.0 (#214) @jameslamb
• Complete Migration of IVF Helpers / Features from RAFT (#213) @tarang-jain
• Use verify-alpha-spec hook (#209) @KyleFromNVIDIA
• Fixes for publishing rust package to crates.io (#207) @benfred
• Add rust example (#206) @benfred
• Adding IVF examples (#203) @cjnolet
• Fix compilation error when _CLK_BREAKDOWN is defined in cagra. (#202) @jiangyinzuo
• DOC: update notebook link (#191) @raybellwaves
• Change cagra.build_index to cagra.build (#187) @benfred
• Add python serialization API's for ivf-pq and ivf_flat (#186) @benfred
• resolve dependency-file-generator warning, rapids-build-backend followup (#185) @jameslamb
• Adopt CI/packaging codeowners (#183) @raydouglass
• Scaling workspace resources (#181) @achirkin
• Remove text builds of documentation (#180) @vyasr
• Add refine to the Python and C api's (#175) @benfred
• Allow serialization on streams (#173) @benfred
• Forward-merge branch-24.06 into branch-24.08 (#169) @benfred
• Use rapids-build-backend (#145) @KyleFromNVIDIA
• ANN_BENCH (#130) @achirkin
• Enable random subsampling (#122) @tfeher

v24.06.00

🐛 Bug Fixes

• Fix CAGRA OOM handling (#167) @tfeher
• Pass through raft static CMake var (#111) @vyasr
• Fix rust docs build (#84) @benfred

📖 Documentation

• chore: update Doxyfile (#162) @eltociear
• cuVS docs updates for release (#161) @cjnolet
• update: fix RAFT URL in README (#91) @hurutoriya
• Update the developer's guide with new copyright hook (#81) @KyleFromNVIDIA
• Add lucene-cuvs to integrations section of docs (#73) @cjnolet

🚀 New Features

• Add refine to public API (#154) @lowener
• [FEA] support of prefiltered brute force (#146) @rhdong
• Migrate IVF-Flat from RAFT (#94) @divyegala
• Migrate IVF-PQ from RAFT to cuVS (#86) @lowener

🛠️ Improvements

• Expose serialization to the python / c-api (#164) @benfred
• Select k instantiations (#159) @benfred
• fix devcontainer name for codespaces (#153) @trxcllnt
• Accept host_mdspan for IVF-PQ build and extend (#148) @tfeher
• Add pairwise_distance api's for C, Python and Rust (#142) @benfred
• Changing RAFT_EXPLICT_* to CUVS_EXPLITI_* (#141) @cjnolet
• Speed-up rust build (#138) @benfred
• Removing libraft.so from libcuvs dependencies (#132) @cjnolet
• CAGRA API update and allow async host refinement (#131) @mfoerste4
• Fix rust api docs (#119) @benfred
• Migrate BFKNN from raft (#118) @benfred
• Fix IVF-PQ helper functions (#116) @lowener
• Migrate raft::cluster to cuvs::cluster (#115) @cjnolet
• hide RAFT #pragma deprecation warnings (#114) @trxcllnt
• Enable Warnings as errors in Python tests (#102) @mroeschke
• Remove libnvjitlink dependency. (#97) @bdice
• Migrate to {{ stdlib("c") }} (#93) @hcho3
• update: replace to cuvs from RAFT in PULL_REQUEST_TEMPLATE (#92) @hurutoriya
• Add python and rust bindings for Ivf-Pq (#90) @benfred
• add --rm and --name to devcontainer run args (#89) @trxcllnt
• Update pip devcontainers to UCX v1.15.0 (#88) @trxcllnt
• Remove gtest from dependencies.yaml (#87) @robertmaynard
• Moving and renaming distance namespaces from raft -> cuvs (#85) @cjnolet
• Use static gtest (#83) @robertmaynard
• Add python and rust bindings for Ivf-Flat (#82) @benfred
• Forward merge branch-24.04 to branch-24.06 (#80) @benfred
• Update devcontainers to use cuda12.2 (#72) @benfred
• Forward merge branch-24.04 to branch-24.06 (#71) @benfred
• Enable forward-merger ops-bot plugin (#70) @benfred
• Adds missing files to update-version.sh (#69) @AyodeAwe
• Add Cagra-Q compression to the python and rust api's (#68) @benfred
• ConfigureCUDA.cmake now sets CUVS_ prefixed variables (#66) @robertmaynard
• Enable all tests for arm jobs (#63) @galipremsagar

v24.04.00

🐛 Bug Fixes

• Fix for 24.04 release (#75) @raydouglass
• Update pre-commit-hooks to v0.0.3 (#60) @KyleFromNVIDIA
• Fix 'cuvs' conda package to only include Python artifacts and clean build.sh (#40) @dantegd

📖 Documentation

• [fix] link anchor of "Installing cuVS" (#64) @yinze00
• More docs updates for 24.04 (#54) @cjnolet
• CAGRA C example and DlPack docs (#51) @divyegala
• Add Rust API docs (#49) @benfred
• quick start and integrations guides for the docs (#47) @cjnolet
• Docs 2404 api docs (#46) @cjnolet
• API Documentation (#44) @cjnolet

🚀 New Features

• CAGRA Python API (#35) @dantegd

🛠️ Improvements

• Use conda env create --yes instead of --force (#62) @bdice
• Add CAGRA-Q compression (#61) @achirkin
• Add Brute Force KNN to Python and Rust API's (#59) @benfred
• expose cuvsResources_t to python (#58) @benfred
• Remove unused get_*.cmake files. (#57) @bdice
• Add ability to allocate with RMM to the c-api and rust api (#56) @benfred
• Update build.sh to build rust bindings (#53) @benfred
• Add upper bound to prevent usage of NumPy 2 (#52) @bdice
• add rust to conda environments (#50) @benfred
• Replace local copyright check with pre-commit-hooks verify-copyright (#48) @KyleFromNVIDIA
• Remove hard-coding of RAPIDS version where possible (#45) @KyleFromNVIDIA
• Expose exception text to C, Python and Rust API's (#39) @benfred
• Adding necessary codeowner groups (#36) @cjnolet
• Add Rust bindings for CAGRA (#34) @benfred
• KNN bruteforce and IVF methods C/C++ API (#33) @viclafargue