Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings
/aoPublic

Pull requests: pytorch/ao

Author
Filter by author
Loading
Label
Filter by label
Loading
Usealt +click/return to exclude labels
or⇧ +click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobodyLoading
Sort

Pull requests list

Add version 2 support for Float8StaticActivationFloat8WeightConfig CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3509 openedDec 18, 2025 byjerryzh168Loading…
[mxfp8 moe training] add wgrad_with_hp option CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. moe mx topic: new featureUse this tag if this PR adds a new feature
#3508 openedDec 18, 2025 bydanielvegamyhreLoading…
[mxfp8 moe training] test floor vs rceil in 3d quantization benchmarks CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. moe mx topic: not user facingUse this tag if you don't want this PR to show up in release notes
#3507 openedDec 17, 2025 bydanielvegamyhreLoading…
[mxfp8 moe training] update benchmark plots to use new kernel CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. moe mx topic: not user facingUse this tag if you don't want this PR to show up in release notes
#3506 openedDec 17, 2025 bydanielvegamyhreLoading…
[mxfp8 moe training] integrate new cuda kernel for blocked layout for groups along K CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. moe mx topic: not user facingUse this tag if you don't want this PR to show up in release notes
#3505 openedDec 17, 2025 bydanielvegamyhreLoading…
[mxfp8 moe training] add CUDA kernel for per-group conversion of scale factors to blocked layout CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. moe mx topic: not user facingUse this tag if you don't want this PR to show up in release notes
#3504 openedDec 17, 2025 bydanielvegamyhreLoading…
[bc-breaking] deprecate old quantization configs CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: bc-breakingUse this tag if this PR breaks backward compatibility
#3503 openedDec 17, 2025 byvkuzoLoading…
[wip] flux eval CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3502 openedDec 17, 2025 byvkuzoLoading…
Fix NVFP4 QAT mixed precision CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: bug fixUse this tag for PRs that fix bugs
#3501 openedDec 17, 2025 byandrewor14Loading…
[xpu][test] Port test_int8_tensor.py UT file to intel XPU CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3500 openedDec 17, 2025 byzxd1997066Loading…
[mxfp8] support RCEIL in triton_to_mxfp8_dim0 kernel with inline PTX CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. mx topic: new featureUse this tag if this PR adds a new feature
#3498 openedDec 17, 2025 bydanielvegamyhreLoading…
[ZENDNN] Adding support for slice in opaque CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3494 openedDec 16, 2025 byPriyjain-amdLoading…
switch llama test scripts to new version of Int4WeightOnlyConfig CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3493 openedDec 16, 2025 byStonepiaLoading…
add pinned memory support for int8tensor CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: new featureUse this tag if this PR adds a new feature
#3489 openedDec 15, 2025 byliangel-02Loading…
[float8] fix the transpose error CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: bug fixUse this tag for PRs that fix bugs topic: not user facingUse this tag if you don't want this PR to show up in release notes
#3488 openedDec 15, 2025 byStonepia • Draft
Parse device at model release script CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: not user facingUse this tag if you don't want this PR to show up in release notes
#3486 openedDec 14, 2025 bynamgyu-younLoading…
Add A100 run to microbenchmarks ciflow/benchmark CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3485 openedDec 12, 2025 byjainapurva • Draft
[xpu][test] Port more cases in test_nf4.py and test_quant_api.py UT files to intel XPU CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: for developersUse this tag if this PR is mainly developer facing xpuIntel XPU related features
#3475 openedDec 10, 2025 byzxd1997066Loading…
make eval script also handle performance measurement CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: for developersUse this tag if this PR is mainly developer facing
#3473 openedDec 9, 2025 byvkuzoLoading…
enable smoothquant for int8 static tensor CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: improvementUse this tag if this PR is an improvement (doesn't fit into any of the other categories)
#3468 openedDec 8, 2025 byjcaipLoading…
Support HQQ for XPU/NPU int4 quantization CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: improvementUse this tag if this PR is an improvement (doesn't fit into any of the other categories)
#3465 openedDec 8, 2025 bynamgyu-younLoading…
support int8 scaled embedding bag pattern match CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. topic: not user facingUse this tag if you don't want this PR to show up in release notes
#3463 openedDec 8, 2025 byshiyang-wengLoading…
Fix type annotation of _choose_qparams_affine/gguf(): input = None CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3461 openedDec 6, 2025 byFreed-WuLoading…
Fix #3458, correct min_val and max_val CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3460 openedDec 6, 2025 byFreed-WuLoading…
Fix bug of _quantize_affine_no_dtype_cast() when zero_point is None CLA SignedThis label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#3459 openedDec 6, 2025 byFreed-WuLoading…
Previous1345…89
Previous
ProTip!no:milestone will show everything without a milestone.
[8]ページ先頭
©2009-2025 Movatter.jp