Skip to content

Pull requests: google/XNNPACK

New pull request New
194 Open 9,100 Closed
194 Open 9,100 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix BF16 dot emulation for dynamic B
#9595 opened Feb 25, 2026 by copybara-service bot Loading…
add rvv fp16 kernels for f16-gemm, f16-igemm, f16-dwconv, f16-spmm
#9594 opened Feb 25, 2026 by ken-unger Loading…
2
Change fp16 softmax kernels to compute the sum as fp32
#9581 opened Feb 21, 2026 by copybara-service bot Loading…
s32_mul for simd-hvx.h use mpyieo for upper 16 bits.
#9566 opened Feb 17, 2026 by copybara-service bot Loading…
Add scalar int4 reduction kernels
#9560 opened Feb 16, 2026 by copybara-service bot Loading…
Optimize ARM SME2 Dot-Product Kernel
#9531 opened Feb 10, 2026 by kasper0406 Loading…
9
Print the fingerprint id string instead of it's value in error messages.
#9503 opened Feb 6, 2026 by copybara-service bot Loading…
Automated Code Change
#9490 opened Feb 5, 2026 by copybara-service bot Loading…
Switch sme2 job to Docker.
#9435 opened Jan 27, 2026 by copybara-service bot Loading…
Upgrade GitHub Actions for Node 24 compatibility
#9415 opened Jan 23, 2026 by salmanmkc Loading…
3
Check the XNNPack caching strategy version when reloading a cache file.
#9410 opened Jan 22, 2026 by copybara-service bot Loading…
Add a caching strategy version number.
#9409 opened Jan 22, 2026 by copybara-service bot Loading…
a very good change.
#9408 opened Jan 22, 2026 by copybara-service bot Loading…
Add int8xint8 FC XNNPACK kernels for AVXVNNIINT8
#9406 opened Jan 21, 2026 by copybara-service bot Loading…
Relax IEEE compliance for hexagon which uses qfloat
#9384 opened Jan 13, 2026 by copybara-service bot Loading…
(DNS) Simple rewrite
#9383 opened Jan 12, 2026 by copybara-service bot Loading…
Test sum_squared reductions at subgraph level
#9374 opened Jan 9, 2026 by copybara-service bot Loading…
Add graph rewrite to luts for sequences of unary functions.
#9373 opened Jan 9, 2026 by copybara-service bot Loading…
Hexagon redefine HVX_Vector types to match v19_0 compiler types
#9346 opened Dec 23, 2025 by copybara-service bot Loading…
Simplify dot by separating the float and integer paths
#9340 opened Dec 22, 2025 by copybara-service bot Loading…
Update scalar GEMM kernels to use unaligned.h
#9293 opened Dec 12, 2025 by copybara-service bot Loading…
Rewrite transpose and stencil_copy
#9281 opened Dec 9, 2025 by copybara-service bot Loading…
SME1 support for int4 GEMM and GEMMV kernels
#9257 opened Dec 5, 2025 by qti-vaiskv • Draft
3
Use unaligned.h loads for qb4w scalar ksum loads
#9227 opened Dec 2, 2025 by GregoryComer Loading…
2
Force split_n to be at least 64 in dot subgraph.
#9194 opened Nov 21, 2025 by copybara-service bot Loading…
ProTip! Exclude everything labeled bug with -label:bug.