Support 8 bit weights "unpacked" compute mode in MatmulNBits kernel #24959
Support 8 bit weights unpacked mode compute in MatmulNBits kernel
9fbda7f4
Modify error message
e9b8e7fb
hariharans29
changed the title Support 8 bit weights unpacked mode compute in MatmulNBits kernel Support 8 bit weights "unpacked" compute mode in MatmulNBits kernel 294 days ago
Update onnxruntime/contrib_ops/cpu/quantization/matmul_nbits.cc
e3bf04f7
Merge branch 'main' of https://github.com/microsoft/onnxruntime into …
39bbbbd5
Fix lint issues
8b2c6820
Merge branch 'hari/matmulnbits_8_bit_fallback' of https://github.com/…
64f7ac5a
PR feedback
954069e4
Nit
0723529f
edgchen1
dismissed these changes
on 2025-06-11
Update onnxruntime/contrib_ops/cpu/quantization/matmul_nbits.cc
6966ff22
hariharans29
dismissed their stale review
via 6966ff22
288 days ago
edgchen1
approved these changes
on 2025-06-12
hariharans29
deleted the hari/matmulnbits_8_bit_fallback branch 286 days ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub