[MLAS] Add CPU DynamicQuantMatMulFp8 contrib op with MLAS FP8 fallback #28416
Add CPU DynamicQuantMatMulFp8 contrib op with MLAS FP8 fallback
536fb3f4
wording for tile replaced with block
b53b1c5a
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
563dcaa4
cleaning the zp checks from MlasFp8GemmBatch after symmetric quantis…
c5fe8e10
melkap01-Arm
marked this pull request as ready for review 65 days ago
documentation updated for failing build, copilot comment addressed
a258ad8d
Reusable A buffer implemented before gemm, tests covering all fp8 typ…
7fb80e7b
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
864b1a89
redundant lines removed
1966da42
Optimize DynamicQuantMatMulFp8 A quantization
98ea9fff
melkap01-Arm
marked this pull request as ready for review 63 days ago
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
473287ec
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
10c88a71
hariharans29
changed the title Add CPU DynamicQuantMatMulFp8 contrib op with MLAS FP8 fallback [MLAS] Add CPU DynamicQuantMatMulFp8 contrib op with MLAS FP8 fallback 61 days ago
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
e041589c
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
b913105a
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
16efd04b
documentation difference patched
4655e914
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
f3018d92
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
cc27b945
answering copilot comment regarding N==0 case
93592f39
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
cdb66cf0
LHS,RHS block layouts changed, scales adjusted
f5424a9b
melkap01-Arm
marked this pull request as ready for review 55 days ago
review comments addressed, docs patched
3696f367
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
68e20378
copilot comments addressed
5e9f1b93
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
68027e92
Merge branch 'microsoft:main' into fp8_DynamicQuantMatMul_Support
7769b190
enum representation cleaned
183c5f7f
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub