-
Notifications
You must be signed in to change notification settings - Fork 4.3k
Pull requests: facebookresearch/faiss
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Set own_fields after reading Index2Layer sub-quantizer during deserialization
CLA Signed
fb-exported
meta-exported
#5147
opened Apr 24, 2026 by
scsiguy
Contributor
Loading…
Fix memory leak warning for unordered_multimap in IndexIVFFlatDedup Python binding (#1667)
CLA Signed
#5145
opened Apr 24, 2026 by
anishesg
Loading…
Add Metal GPU backend for Apple Silicon (IndexFlat)
CLA Signed
install
Integration
Performance
#5144
opened Apr 24, 2026 by
Evandabest
Loading…
Enforce minimum training set size for ScalarQuantizer, check nullptr (#5141)
CLA Signed
fb-exported
meta-exported
#5141
opened Apr 23, 2026 by
ddrcoder
Contributor
Loading…
Enable dynamic dispatch build on Windows (iteration on #5127)
CLA Signed
#5137
opened Apr 22, 2026 by
mdouze
Contributor
Loading…
1 task
Introduce early stop facilities for IVF
CLA Signed
#5135
opened Apr 21, 2026 by
alexanderguzhva
Contributor
Loading…
Add faiss-gpu pip wheel packaging (#5131)
CLA Signed
fb-exported
meta-exported
#5131
opened Apr 20, 2026 by
alibeklfc
Contributor
Loading…
introduce RVV platform and its ScalarQuantizer
CLA Signed
#5128
opened Apr 20, 2026 by
alexanderguzhva
Contributor
Loading…
Work around GCC 12 miscompilation of AVX2 histogram (#5124)
CLA Signed
fb-exported
meta-exported
#5124
opened Apr 20, 2026 by
mdouze
Contributor
Loading…
Reuse per-thread query buffers in IVFRaBitQFastScan compute_LUT
CLA Signed
#5123
opened Apr 19, 2026 by
lyang24
Contributor
Loading…
Fix peak-memory spike when loading IVF invlists via IO_FLAG_MMAP_IFC
CLA Signed
#5122
opened Apr 19, 2026 by
yantis
Loading…
2 of 3 tasks
fix: use safe deserialization in test_read_index_deserialize.cpp
CLA Signed
#5119
opened Apr 18, 2026 by
orbisai0security
Loading…
3 tasks done
Make BLAS threshold scale with OMP thread count (#5109)
CLA Signed
fb-exported
meta-exported
#5109
opened Apr 16, 2026 by
mnorris11
Contributor
Loading…
RaBitQ: simplify compute_query_factors preprocessing hot path
CLA Signed
#5103
opened Apr 15, 2026 by
lyang24
Contributor
Loading…
IVFRaBitQFastScan: add refine_error_scale to trade a small amount of recall for higher nbits=2 QPS
CLA Signed
#5101
opened Apr 15, 2026 by
lyang24
Contributor
Loading…
ARM SVE optimization for L2 distance kernels adapted to new simd_impl structure
CLA Signed
#5098
opened Apr 14, 2026 by
ThatikondV
Loading…
Add explicit Rule of Five defaults for classes with user-provided destructors (#5094)
CLA Signed
fb-exported
meta-exported
#5094
opened Apr 13, 2026 by
junjieqi
Contributor
Loading…
Suppress non-const global variable warnings and fix security integer casts (#5092)
CLA Signed
fb-exported
meta-exported
#5092
opened Apr 13, 2026 by
junjieqi
Contributor
Loading…
Add braces around single-line control flow statements (#5090)
CLA Signed
fb-exported
meta-exported
#5090
opened Apr 13, 2026 by
junjieqi
Contributor
Loading…
Convert binary index Hamming callers to dynamic dispatch (#5071)
CLA Signed
fb-exported
ignore
meta-exported
#5071
opened Apr 9, 2026 by
algoriddle
Contributor
Loading…
Convert hamming.cpp public API to dynamic dispatch (#5070)
CLA Signed
fb-exported
ignore
meta-exported
#5070
opened Apr 9, 2026 by
algoriddle
Contributor
Loading…
Using AVX-512 VNNI instructions to optimize the DistanceComputerByte::compute_code_distance
CLA Signed
#5067
opened Apr 9, 2026 by
baohongmin
Loading…
Fix: static SIMD dispatch falls to scalar for avx512_spr/avx512/arm_sve builds
CLA Signed
#5057
opened Apr 7, 2026 by
mulugetam
Contributor
Loading…
Fix Windows ARM64 (MSVC) build: disable NEON SIMD path
CLA Signed
#5046
opened Apr 6, 2026 by
Chessing234
Contributor
Loading…
3 tasks
Previous Next
ProTip!
What’s not been updated in a month: updated:<2026-03-25.