transformers
[Performance] FP8 Grouped and Batched Matmuls
#44231
Merged

[Performance] FP8 Grouped and Batched Matmuls #44231

ArthurZucker merged 27 commits into main from fp8-grouped-mm
IlyasMoutawwakil
IlyasMoutawwakil simplify
1984e5da
IlyasMoutawwakil finegrained fp8 moe forwards
b1fcbd80
IlyasMoutawwakil optimized fp8 fused, batched and grouped paths
12b05465
IlyasMoutawwakil Merge branch 'main' into fp8-grouped-mm
f47040fe
HuggingFaceDocBuilderDev
IlyasMoutawwakil fix
84e9ef21
IlyasMoutawwakil wrap triton
94e4cd79
IlyasMoutawwakil fix calls
98475580
IlyasMoutawwakil fix
2aa637b5
IlyasMoutawwakil Merge branch 'main' into fp8-grouped-mm
57e47798
IlyasMoutawwakil remove fused quant kernel (litlle gain and unnecessary) and use torch…
125d8f4e
IlyasMoutawwakil use kernels
a2e7dd12
IlyasMoutawwakil fix
71a1b8c2
IlyasMoutawwakil no need to wrap cutlass
5c33299d
IlyasMoutawwakil
IlyasMoutawwakil commented on 2026-02-26
IlyasMoutawwakil cleanup
9212cc37
IlyasMoutawwakil fix
ffe79316
IlyasMoutawwakil Merge branch 'main' into fp8-grouped-mm
3b9e9f6c
IlyasMoutawwakil Merge branch 'main' into fp8-grouped-mm
fef6f359
IlyasMoutawwakil added non gated experts support
25aedb2c
IlyasMoutawwakil remove comments
7e7e2ac7
IlyasMoutawwakil style
6c6e1768
IlyasMoutawwakil fix
4ab554db
IlyasMoutawwakil IlyasMoutawwakil requested a review from copilot-pull-request-reviewer copilot-pull-request-reviewer 106 days ago
copilot-pull-request-reviewer
copilot-pull-request-reviewer commented on 2026-03-03
IlyasMoutawwakil Update src/transformers/quantizers/quantizer_finegrained_fp8.py
8243a429
IlyasMoutawwakil Update finegrained_fp8.py
77dde4e6
IlyasMoutawwakil IlyasMoutawwakil marked this pull request as ready for review 105 days ago
IlyasMoutawwakil per tensor scaling support
3802cd43
IlyasMoutawwakil IlyasMoutawwakil requested a review from SunMarc SunMarc 104 days ago
IlyasMoutawwakil IlyasMoutawwakil requested a review from ArthurZucker ArthurZucker 104 days ago
IlyasMoutawwakil IlyasMoutawwakil requested a review from Cyrilvallez Cyrilvallez 103 days ago
IlyasMoutawwakil IlyasMoutawwakil requested a review from vasqu vasqu 103 days ago
SunMarc
SunMarc approved these changes on 2026-03-05
Cyrilvallez
Cyrilvallez approved these changes on 2026-03-09
ArthurZucker
ArthurZucker approved these changes on 2026-03-10
IlyasMoutawwakil use custom fp8 interface
6fa940f0
IlyasMoutawwakil document
eca2f01b
github-actions
SunMarc Merge branch 'main' into fp8-grouped-mm
c3107a90
SunMarc
SunMarc approved these changes on 2026-03-10
SunMarc SunMarc enabled auto-merge 98 days ago
ArthurZucker ArthurZucker merged ff2ba441 into main 98 days ago
ArthurZucker ArthurZucker deleted the fp8-grouped-mm branch 98 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone