PR #25582 Enable 2bit CPU matmul fallback

Enable 2bit CPU matmul fallback #25582

carzh merged 11 commits into main from carzh/enable-2bit-fallback

enable fallback + switch to nbits quantizer

90f7c26e

fix for quantizer script

e970f745

figured out bug so uncommenting enforce line

ef145419

HectorSVC commented on 2025-07-29

2bit quant script fix

69adef03

added working matmul 2bits unit test

b591647b

github-actions commented on 2025-07-30

github-advanced-security commented on 2025-07-30

carzh marked this pull request as ready for review 201 days ago

cleanup + lintrunner sigh

02c9ae39

test wip for quantizer python

c38e8a45

github-advanced-security commented on 2025-07-30

hariharans29 commented on 2025-07-30

fixed 4bit matmulnbitsquantizer.py test, applied suggestions (fixed c…

c4cf9b14

HectorSVC commented on 2025-08-01

updated test_op_matmul_2bits.py

f1d6470c

removed gather tests as well

7246934c

github-actions commented on 2025-08-01

github-advanced-security commented on 2025-08-01

format

6f10276c

HectorSVC approved these changes on 2025-08-04

hariharans29 approved these changes on 2025-08-04

carzh merged d5d3b287 into main 196 days ago

carzh deleted the carzh/enable-2bit-fallback branch 196 days ago

Reviewers

hariharans29

HectorSVC

github-actions

github-advanced-security

Assignees

No one assigned

Labels

None yet

Milestone

No milestone

onnxruntime Enable 2bit CPU matmul fallback #25582 Merged

Enable 2bit CPU matmul fallback #25582

onnxruntime
Enable 2bit CPU matmul fallback
#25582

Merged