Add float zero point support for 2-bit LUT GEMM in MatMulNBits #28354
Add float zero point support for 2-bit LUT GEMM in MatMulNBits
e4c4922c
Address review: add BF16 guard, LUT skip, fallback test
7aaf31ad
Fix doc comment: clarify float ZP shape vs packer usage
4e39a916
Fix test packed stride, extract ZP conversion helper, add non-aligned…
b99dc2f4
vraspar
marked this pull request as ready for review 45 days ago
Address review feedback for float ZP LUT GEMM support
004beb1e
Add varying per-block and dynamic ZP test coverage for 2-bit float ze…
6c6f3943
Merge remote-tracking branch 'origin/main' into pr-28354
6182e93a
Update QMoE LutGemmPack calls for widened float ZP API
3047db86
tianleiwu
approved these changes
on 2026-05-13
vraspar
enabled auto-merge (squash) 38 days ago
vraspar
merged
e8ae6ce9
into main 38 days ago
vraspar
deleted the vraspar/matmulnbits-float-zp branch 38 days ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub