Skip to content

Pull requests: ROCm/TransformerEngine

New pull request New
17 Open 344 Closed
17 Open 344 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Cherry picking keep_fp8_weight_transpose_cache flag refactor and fsdp2 fp8 autocast all gather commits
#389 opened Dec 2, 2025 by sudhu2k Loading…
9 of 13 tasks
6
Current scaling: two-stage Triton amax kernel
#385 opened Nov 26, 2025 by matthiasdiener Loading…
6 of 13 tasks
@matthiasdiener
2
Enable AOTriton BWD V3 API
#382 opened Nov 25, 2025 by Micky774 Loading…
13 tasks
@Micky774
6
Old FP8 support code cleanup
#379 opened Nov 24, 2025 by ipanfilo Loading…
1 of 13 tasks
Re-enable supported GEMM configs
#378 opened Nov 24, 2025 by ipanfilo Loading…
13 tasks
Layernorm forward optimization
#377 opened Nov 24, 2025 by eliotwang Loading…
13 tasks
IFU dev v2.6
#374 opened Nov 19, 2025 by wangye805 Loading…
9 of 13 tasks
94
Userbuffer epic
#367 opened Nov 11, 2025 by alextmagro • Draft
AITER Native Padding Support and BSHD + Padding --> THD + Padding conversion
#354 opened Oct 28, 2025 by Micky774 Loading…
13 tasks
55
JAX FA Benchmarking Script
#351 opened Oct 24, 2025 by Micky774 Loading…
13 tasks
@Micky774
18
Fused Cross Entropy Triton - Loss Scaling and Vanishing Grads Bugfix
#336 opened Oct 16, 2025 by sarthak-amd Loading…
1 task done
7
Triton norms dispatch refactor
#305 opened Sep 5, 2025 by Micky774 Loading…
13 tasks
12
heyi's layernorm optimization
#225 opened Jul 3, 2025 by eliotwang Loading…
8 of 13 tasks
@charyang-ai
Added Dockerfile for CI images
#195 opened May 28, 2025 by VeeraRajasekhar Loading…
7 of 13 tasks
30
[ROCm] support triton-based flash-attn in TE
#177 opened May 1, 2025 by wangye805 Loading…
8 of 13 tasks
6
Update attention example attention.ipynb
#152 opened Mar 19, 2025 by anhminhnguyenhoang Loading…
5 of 13 tasks
1
Honor the NVTE_FUSED_ATTN_<backend> in test_fused_attn.py
#123 opened Feb 11, 2025 by wangye805 Loading…
13 tasks
3
ProTip! no:milestone will show everything without a milestone.