Skip to content

Weijiac/dsv4 bridge#4770

Draft
weijiac0619 wants to merge 436 commits into
NVIDIA:mainfrom
weijiac0619:weijiac/dsv4-bridge
Draft

Weijiac/dsv4 bridge#4770
weijiac0619 wants to merge 436 commits into
NVIDIA:mainfrom
weijiac0619:weijiac/dsv4-bridge

Conversation

@weijiac0619
Copy link
Copy Markdown
Contributor

@weijiac0619 weijiac0619 commented May 13, 2026

What does this PR do ?

⚠️ For major changes (either in lines of code or in its impact), please make sure to first share a design doc with the team. If you're unsure what's the best way to do so, contact the @mcore-oncall.

Issue tracking

For PRs from open-source community contributors:

  • New features: a linked issue is required. Please open a feature request and reference it here before submitting the PR.
  • Small updates (bug fixes, minor improvements): a linked issue is recommended and will accelerate the PR review process.

Linked issue:

Contribution process

Pre-checks

  • I have added relevant unit tests
  • I have added relevant functional tests
  • I have added proper typing to my code Typing guidelines
  • I have added relevant documentation
  • I have run the autoformatter.sh on my PR

Code review

Feel free to message or comment the @mcore-oncall to help accelerate your merge into main. The less complex your PR is, the faster it will be approved and merged!

All PRs start as draft. If you open a non-draft PR, it will be automatically converted to draft.

Step 1: Mark PR as "Ready for Review"

  1. When your PR is ready, click Ready for Review.
  2. An oncall reviewer is auto-assigned and expert reviewers are notified based on your changes.
    • Some PRs may jump straight to step 2. This is determined by .github/CODEOWNERS.

⚠️ Only mark as ready once merge-conflicts are resolved and the CI is passing.
Final Review might get declined if these requirements are not fulfilled.

Step 2: Final Review

For PRs that change megatron/core, once all expert reviewers have approved, the Final Review label is applied automatically and final reviewers are assigned.

For PRs outside megatron/core, this step is skipped.

Step 3: Approved

Once all required reviewers have approved, the Approved label is applied automatically.

Merge

Any member of mcore-engineers will be able to merge your PR.

For MRs into `dev` branch The proposed review process for `dev` branch is under active discussion.

MRs are mergable after one approval by either eharper@nvidia.com or zijiey@nvidia.com.

pablo-garay and others added 30 commits December 4, 2025 15:57
@pablo-garay
update API compat check baseline to 274e04d (NVIDIA#2548)
87ac13d 
Signed-off-by: Pablo Garay <pagaray@nvidia.com>
@pablo-garay
feat: mcore trigger mbridge (NVIDIA#2340) (NVIDIA#2552)
f0c1b55 
Signed-off-by: Pablo Garay <pagaray@nvidia.com>
@buptzyb
[Dev] Optimize TE CUDA Graph capturing time (NVIDIA#2483)
8de5a7f 
Signed-off-by: Robin Zhang <robinz@nvidia.com>
@Jianbing-D @shjwudp
@FDecaYed @kevalmorabia97 @yashaswikarnati @jaredcasper @asolergi-nv @ko3n1g @ananthsub @tdene @sidsingh-nvidia @dimapihtar @youngeunkwon0405 @lmcafee-nvidia @maanug-nv @ajschmidt8 @deepakn94 @mathemakitten @github-actions @Skylion007 @ArEsKay3 @yeyu-nvidia @aklife97 @rogerwaleffe @chtruong814 @Autumn1998 @zhongbozhu @xrennvidia @yaox12 @yanring @santhnm2 @pablo-garay @AAnoosheh @kanz-nv @jalbericiola @jon-barker @cuichenx @theothermike @BestJuly @sbhavani @Li-dongyang @buptzyb @kunlunl
[Dev] Feature: linear cross entropy fusion (NVIDIA#2256)
1f08ceb 
Signed-off-by: Jianbing Dong <jianbingd@nvidia.com>
Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Signed-off-by: oliver könig <okoenig@nvidia.com>
Signed-off-by: Ananth Subramaniam <ansubramania@nvidia.com>
Signed-off-by: dimapihtar <dpihtar@gmail.com>
Signed-off-by: Youngeun Kwon <youngeunk@nvidia.com>
Signed-off-by: Youngeun <kyeg9404@gmail.com>
Signed-off-by: Maanu Grover <maanug@nvidia.com>
Signed-off-by: ykarnati <ykarnati@nvidia.com>
Signed-off-by: Deepak Narayanan <dnarayanan@nvidia.com>
Signed-off-by: GitHub Actions <github-actions[bot]@users.noreply.github.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Zhongbo Zhu <zhongboz@nvidia.com>
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Signed-off-by: Xin Yao <xiny@nvidia.com>
Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>
Signed-off-by: Pablo Garay <pagaray@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Chen Cui <chcui@nvidia.com>
Signed-off-by: Li Tao <lit@nvidia.com>
Signed-off-by: lit <lit@nvidia.com>
Signed-off-by: Santosh Bhavani <santosh.bhavani@live.com>
Signed-off-by: Robin Zhang <robinz@nvidia.com>
Signed-off-by: kunlunl <kunlunl@nvidia.com>
Co-authored-by: Jianbin Chang <shjwudp@gmail.com>
Co-authored-by: Deyu Fu <Deyu.Foo@gmail.com>
Co-authored-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Co-authored-by: Yashaswi Karnati <144376261+yashaswikarnati@users.noreply.github.com>
Co-authored-by: Jared Casper <155158+jaredcasper@users.noreply.github.com>
Co-authored-by: Antoni-Joan Solergibert <asolergibert@nvidia.com>
Co-authored-by: oliver könig <okoenig@nvidia.com>
Co-authored-by: Ananth Subramaniam <ansubramania@nvidia.com>
Co-authored-by: Teodor-Dumitru Ene <34819528+tdene@users.noreply.github.com>
Co-authored-by: Siddharth Singh <136645615+sidsingh-nvidia@users.noreply.github.com>
Co-authored-by: Mcore Bot <mcore-bot@nvidia.com>
Co-authored-by: Dmytro Pykhtar <37850217+dimapihtar@users.noreply.github.com>
Co-authored-by: Youngeun Kwon <youngeunk@nvidia.com>
Co-authored-by: Lawrence McAfee <85179052+lmcafee-nvidia@users.noreply.github.com>
Co-authored-by: Maanu Grover <109391026+maanug-nv@users.noreply.github.com>
Co-authored-by: Lawrence McAfee <lmcafee@nvidia.com>
Co-authored-by: AJ Schmidt <ajschmidt8@users.noreply.github.com>
Co-authored-by: Deepak Narayanan <2724038+deepakn94@users.noreply.github.com>
Co-authored-by: helen ngo <helenn@nvidia.com>
Co-authored-by: GitHub Actions <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com>
Co-authored-by: Robert Kirby <rkirby@nvidia.com>
Co-authored-by: Teodor-Dumitru Ene <tene@nvidia.com>
Co-authored-by: yeyu-nvidia <yeyu@nvidia.com>
Co-authored-by: Abhinav Khattar <akhattar@nvidia.com>
Co-authored-by: Roger Waleffe <rwaleffe@nvidia.com>
Co-authored-by: Charlie Truong <chtruong@nvidia.com>
Co-authored-by: Tong Liu <liutongt1998@gmail.com>
Co-authored-by: Zhongbo Zhu <42691305+zhongbozhu@users.noreply.github.com>
Co-authored-by: Xiaowei Ren <xren@nvidia.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
Co-authored-by: Teodor-Dumitru Ene <teodord.ene@gmail.com>
Co-authored-by: Zijie Yan <zijiey@nvidia.com>
Co-authored-by: root <root@pool0-01101.cm.cluster>
Co-authored-by: Keshav Santhanam <ksanthanam@nvidia.com>
Co-authored-by: Pablo Garay <pagaray@nvidia.com>
Co-authored-by: Asha Anoosheh <aanoosheh@nvidia.com>
Co-authored-by: Kan Zhu <kanz@nvidia.com>
Co-authored-by: Robert Kirby <rkirby@cw-dfw-cs-001-vscode-01.cm.cluster>
Co-authored-by: Jorge Albericio <jalbericiola@nvidia.com>
Co-authored-by: Jon Barker <19699370+jon-barker@users.noreply.github.com>
Co-authored-by: Chen Cui <chcui@nvidia.com>
Co-authored-by: Pablo Garay <palenq@gmail.com>
Co-authored-by: Tong Liu <tongliu@nvidia.com>
Co-authored-by: Michael Wojcikiewicz <mwojcikiewic@nvidia.com>
Co-authored-by: Li Tao <lit@nvidia.com>
Co-authored-by: Santosh Bhavani <santosh.bhavani@live.com>
Co-authored-by: Li Ruixiao <cgruixiao@outlook.com>
Co-authored-by: Robin Zhang <robinz@nvidia.com>
Co-authored-by: Kunlun Li <94586211+kunlunl@users.noreply.github.com>
@yuzhongw-nvidia @kunlunl
Fix gpt_layer_spec for frequently linear attention (NVIDIA#2481)
9cf6838 
Co-authored-by: Kunlun Li <94586211+kunlunl@users.noreply.github.com>
@Niccolo-Ajroldi
Skip trainloader when args.skip_train is True (NVIDIA#2501)
89fe895
@FDecaYed
[DEV] fixes for muon(qwen3-next, ep multi-adam) (NVIDIA#2564)
a6d86a6 
Signed-off-by: Deyu Fu <deyuf@nvidia.com>
@HaochenYuan
[Dev] remove fp16 assert in moe_grouped_gemm & EP (NVIDIA#2494)
aee4a74
@skyw
Update tp support in muon (NVIDIA#2385)
dfe4da2 
Signed-off-by: Hao Wu <skyw@nvidia.com>
@Victarry
[DEV] Update GitHub MoE functional test cases (NVIDIA#2449)
1d462bd
@rj42 @yaox12
Fix: don't enter branch if mtp_num_layers == 0 (NVIDIA#2581)
23e092f 
Co-authored-by: Xin Yao <xiny@nvidia.com>
@lhb8125 @yanring @Wohox
[Dev] fix(moe): Support HybridEP and reduce memory overhead for 1F1B …
c60d5c2 
…A2A overlap (NVIDIA#2201)

Signed-off-by: Hongbin Liu <hongbinl@nvidia.com>
Signed-off-by: Pingtian Li <pingtianl@nvidia.com>
Co-authored-by: root <root@eos0318.eos.clusters.nvidia.com>
Co-authored-by: Zijie Yan <zijiey@nvidia.com>
Co-authored-by: Pingtian Li <pingtianl@nvidia.com>
@FDecaYed
Merge branch 'main' into dev
4db2f11 
Signed-off-by: Deyu Fu <deyuf@nvidia.com>
@ko3n1g @kevalmorabia97
@yashaswikarnati @jaredcasper @asolergi-nv @shjwudp @ananthsub @tdene @sidsingh-nvidia @dimapihtar @youngeunkwon0405 @lmcafee-nvidia @maanug-nv @ajschmidt8 @deepakn94 @mathemakitten @github-actions @Skylion007 @ArEsKay3 @yeyu-nvidia @aklife97 @rogerwaleffe @chtruong814 @Autumn1998 @zhongbozhu @xrennvidia @yaox12 @yanring @santhnm2 @pablo-garay @AAnoosheh @kanz-nv @jalbericiola @jon-barker @cuichenx @Victarry @yobibyte @theothermike @shanmugamr1992 @cspades @guyueh1
[dev] pull main 1201 (NVIDIA#2448)
ed804b4 
Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Signed-off-by: oliver könig <okoenig@nvidia.com>
Signed-off-by: Ananth Subramaniam <ansubramania@nvidia.com>
Signed-off-by: dimapihtar <dpihtar@gmail.com>
Signed-off-by: Youngeun Kwon <youngeunk@nvidia.com>
Signed-off-by: Youngeun <kyeg9404@gmail.com>
Signed-off-by: Maanu Grover <maanug@nvidia.com>
Signed-off-by: ykarnati <ykarnati@nvidia.com>
Signed-off-by: Deepak Narayanan <dnarayanan@nvidia.com>
Signed-off-by: GitHub Actions <github-actions[bot]@users.noreply.github.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Zhongbo Zhu <zhongboz@nvidia.com>
Signed-off-by: Xiaowei Ren <xren@nvidia.com>
Signed-off-by: Xin Yao <xiny@nvidia.com>
Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>
Signed-off-by: Pablo Garay <pagaray@nvidia.com>
Signed-off-by: Asha Anoosheh <aanoosheh@nvidia.com>
Signed-off-by: Chen Cui <chcui@nvidia.com>
Signed-off-by: Cory Ye <cye@nvidia.com>
Signed-off-by: Guyue Huang <guyueh@nvidia.com>
Signed-off-by: Deyu Fu <deyuf@nvidia.com>
Co-authored-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Co-authored-by: Yashaswi Karnati <144376261+yashaswikarnati@users.noreply.github.com>
Co-authored-by: Jared Casper <155158+jaredcasper@users.noreply.github.com>
Co-authored-by: Antoni-Joan Solergibert <asolergibert@nvidia.com>
Co-authored-by: Jianbin Chang <shjwudp@gmail.com>
Co-authored-by: oliver könig <okoenig@nvidia.com>
Co-authored-by: Ananth Subramaniam <ansubramania@nvidia.com>
Co-authored-by: Teodor-Dumitru Ene <34819528+tdene@users.noreply.github.com>
Co-authored-by: Siddharth Singh <136645615+sidsingh-nvidia@users.noreply.github.com>
Co-authored-by: Mcore Bot <mcore-bot@nvidia.com>
Co-authored-by: Dmytro Pykhtar <37850217+dimapihtar@users.noreply.github.com>
Co-authored-by: Youngeun Kwon <youngeunk@nvidia.com>
Co-authored-by: Lawrence McAfee <85179052+lmcafee-nvidia@users.noreply.github.com>
Co-authored-by: Maanu Grover <109391026+maanug-nv@users.noreply.github.com>
Co-authored-by: Lawrence McAfee <lmcafee@nvidia.com>
Co-authored-by: AJ Schmidt <ajschmidt8@users.noreply.github.com>
Co-authored-by: Deepak Narayanan <2724038+deepakn94@users.noreply.github.com>
Co-authored-by: helen ngo <helenn@nvidia.com>
Co-authored-by: GitHub Actions <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com>
Co-authored-by: Robert Kirby <rkirby@nvidia.com>
Co-authored-by: Teodor-Dumitru Ene <tene@nvidia.com>
Co-authored-by: yeyu-nvidia <yeyu@nvidia.com>
Co-authored-by: Abhinav Khattar <akhattar@nvidia.com>
Co-authored-by: Roger Waleffe <rwaleffe@nvidia.com>
Co-authored-by: Charlie Truong <chtruong@nvidia.com>
Co-authored-by: Tong Liu <liutongt1998@gmail.com>
Co-authored-by: Zhongbo Zhu <42691305+zhongbozhu@users.noreply.github.com>
Co-authored-by: Xiaowei Ren <xren@nvidia.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
Co-authored-by: Teodor-Dumitru Ene <teodord.ene@gmail.com>
Co-authored-by: Zijie Yan <zijiey@nvidia.com>
Co-authored-by: root <root@pool0-01101.cm.cluster>
Co-authored-by: Keshav Santhanam <ksanthanam@nvidia.com>
Co-authored-by: Pablo Garay <pagaray@nvidia.com>
Co-authored-by: Asha Anoosheh <aanoosheh@nvidia.com>
Co-authored-by: Kan Zhu <kanz@nvidia.com>
Co-authored-by: Robert Kirby <rkirby@cw-dfw-cs-001-vscode-01.cm.cluster>
Co-authored-by: Jorge Albericio <jalbericiola@nvidia.com>
Co-authored-by: Jon Barker <19699370+jon-barker@users.noreply.github.com>
Co-authored-by: Chen Cui <chcui@nvidia.com>
Co-authored-by: Pablo Garay <palenq@gmail.com>
Co-authored-by: Tong Liu <tongliu@nvidia.com>
Co-authored-by: Dennis(Zhenhuan) Liu <denliu@nvidia.com>
Co-authored-by: yobi byte <yobibyte@users.noreply.github.com>
Co-authored-by: Jon Barker <jbarker@cw-dfw-cs-001-vscode-01.cm.cluster>
Co-authored-by: Michael Wojcikiewicz <mwojcikiewic@nvidia.com>
Co-authored-by: Shanmugam Ramasamy <111910568+shanmugamr1992@users.noreply.github.com>
Co-authored-by: Shanmugam Ramasamy <shanmugamr@cw-dfw-cs-001-login-02.cm.cluster>
Co-authored-by: Shanmugam Ramasamy <shanmugamr@cw-dfw-cs-001-login-01.cm.cluster>
Co-authored-by: Cory Ye <44509866+cspades@users.noreply.github.com>
Co-authored-by: Guyue Huang <140554423+guyueh1@users.noreply.github.com>
@ko3n1g
chore: Bump baseline (NVIDIA#2626)
2d398b4 
Signed-off-by: oliver könig <okoenig@nvidia.com>
@Autumn1998
[Dev] Use the latest Hybrid-EP (NVIDIA#2424)
e8a9275
@pablo-garay
API compat: ignore ParameterMovedBreakage for __init__ methods (NVIDI…
305957a 
…A#2649)

Signed-off-by: Pablo Garay <pagaray@nvidia.com>
@maanug-nv @ericharper
[training migration] add training config dataclass and arg generation…
e93814b 
… utility (NVIDIA#2651)

Signed-off-by: Maanu Grover <maanug@nvidia.com>
Co-authored-by: Eric Harper <eharper@nvidia.com>
@buptzyb
[Dev] Optimize TE CUDA Graph _get_sample_arguments() Time (NVIDIA#2568)
288b8ea 
Signed-off-by: Robin Zhang <robinz@nvidia.com>
@yuzhongw-nvidia @ko3n1g
Reopen qwen3next functional test in lightweight mode (NVIDIA#2493)
0eec631 
Signed-off-by: oliver könig <okoenig@nvidia.com>
Co-authored-by: oliver könig <okoenig@nvidia.com>
@buptzyb
[Dev] Fix CUDA RNG Tracker (NVIDIA#2640)
2ebff67 
Signed-off-by: Robin Zhang <robinz@nvidia.com>
@pablo-garay
[Dev] Mark API backwards compatibility checks as OPTIONAL (non-blocki…
368e580 
…ng) (NVIDIA#2699)
@kunlunl @shjwudp
[Dev] FP8 params support for megatron-fsdp (MXFP8/Blockwise) (NVIDIA#…
3714d81 
…2086)

Signed-off-by: kunlunl <kunlunl@nvidia.com>
Co-authored-by: jianbinc <shjwudp@gmail.com>
@yuzhongw-nvidia
[Dev] Feat(moe): Gated delta net context parallel (CP) (NVIDIA#2614)
a935008
@ko3n1g @Victarry
ci: Gridify test configs (NVIDIA#2707)
fd932c9 
Signed-off-by: oliver könig <okoenig@nvidia.com>
Co-authored-by: Dennis Liu <denliu@nvidia.com>
@ko3n1g
Revert "[dev] Add assertion for mxfp8 params without dp overlap (NVID…
2b1fc70 
…IA#2270)"

This reverts commit 7968d5f.
@ko3n1g
Revert "[Dev] Use the latest Hybrid-EP (NVIDIA#2424)" (NVIDIA#2732)
4665be4
@Wohox
[Dev] Fix ep overlap missing final layernorm (NVIDIA#2691)
46b5505
@HaochenYuan @BestJuly
[Dev] Remove calculation of padding token in moe routing loss (NVIDIA…
0b6714e 
…#2121)

Co-authored-by: Li Tao <lit@nvidia.com>
@chtruong814
Revert "[Dev] Remove calculation of padding token in moe routing loss (…
1068d77 
…NVIDIA#2121)" (NVIDIA#2747)

Signed-off-by: Charlie Truong <chtruong@nvidia.com>
rapatel and others added 29 commits April 17, 2026 10:59
@rapatel @lhb8125 @yaox12
Allow fine-grained offloading with MC impl of full-CG. (NVIDIA#4253)
6efb083 
Co-authored-by: Hongbin Liu <lhb8125@users.noreply.github.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
@sraman-rgb @yaox12
Add TEFusedDenseMLP for Dense+Grouped GEMM fusion on SM100+ (NVIDIA#4318
be3b874 
)

Co-authored-by: Siddhartha Raman S <sraman@login-lyris02.lyris.clusters.nvidia.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
@Victarry @claude
[Dev] Fix docs build from main sync (NVIDIA#4356)
1b47bc0 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@nanz-nv
[Dev] overload factor logging (NVIDIA#4110)
55df4e5
@wplf
Revert "fix mfsdp unwrap stuck at MegatronFSDP [dev] (NVIDIA#4273)" (N…
13557a2 
…VIDIA#4393)
@Phlip79 @shifangx @yaox12
M4 leftover for TE cuda graph (dev) (NVIDIA#4369)
546a448 
Co-authored-by: Shifang Xu <shifangx@nvidia.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
@gdengk
[Dev] Add high-priority a2a comm stream option and hybridep preproces…
85bced0 
…sing SMs (NVIDIA#4401)

Co-authored-by: Gao Deng <gdeng@login-lyris02.lyris.clusters.nvidia.com>
@github-actions
Merge remote-tracking branch 'origin/main' into main2dev/22_04_2026
57005c8
@github-actions @claude
chore: post-merge fixes for nightly sync main into dev (22_04_2026)
8add4e4 
- Restore dev's pyproject.toml, uv.lock, and Dockerfile.ci.dev
- Update nvidia-resiliency-ext to main's revision (required for get_write_results_queue)
- Fix hybrid_model.py: init_chunk_handler() missing pp_rank, delta_offload_bytes_across_pp_ranks, activation_offload_fraction params
- Fix hybrid_model.py: mark_not_offloadable() -> mark_not_offload()
- Run black + isort on all changed Python files

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@github-actions @claude
fix: revert nvidia-resiliency-ext revision to match uv.lock
baa3df4 
Restore dev's nvidia-resiliency-ext revision to keep pyproject.toml
and uv.lock consistent. The mismatch caused uv sync --locked to fail
in CI linting. The get_write_results_queue import in torch.py is a
lazy runtime import that won't be hit during linting or unit tests.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@github-actions @claude
fix: reformat 4 files with correct black==24.4.2 and isort==5.13.2
bbb06e2 
Previous formatting used wrong tool versions (black 24.10.0, isort 8.0.1).
Re-ran with CI-pinned versions: black==24.4.2, isort==5.13.2.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@svcnvidia-nemo-ci @claude
fix: restore missing ArgumentGroupFactory import in arguments.py
89798f3 
The merge removed the import of ArgumentGroupFactory from
argument_utils but it is still used extensively in the file.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@gdengk
Reorder mtp_post_process after attn backward in 1F1B schedule plan (N…
db5ade5 
…VIDIA#4430)

Co-authored-by: Gao Deng <gdeng@login-lyris01.lyris.clusters.nvidia.com>
@Phlip79 @claude
chore: keep CODEOWNERS unchanged in main→dev sync
8cf5458 
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@Phlip79 @claude
chore: update gpt3_mcore_te_tp2_pp2_mhc golden values for main→dev sync
fba3a80 
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@cspades
[Dev] Fix mis-set decoupled gradient for Megatron-FSDP. (NVIDIA#4426)
78858b2 
Signed-off-by: Cory Ye <cye@nvidia.com>
@ko3n1g @Victarry
@claude @Phlip79 @yaox12 @cspades @dimapihtar @maanug-nv @claude @BestJuly @github-actions @janEbert @shifangx @ZhiyuLi-Nvidia @ezioliao @ericharper @deepakn94 @tdene @yuzhongw-nvidia @DhineshPonnarasan @sbak5 @chtruong814 @kunlunl @wdykas @Mellonta @santhnm2 @awsankur @asolergi-nv @jon-barker @megnvidia @mkhona-nvidia @wujingyue @jalbericiola @sbhavani @youngeunkwon0405
chore: nightly sync main into dev (22_04_2026) (NVIDIA#4436)
64d2e0a 
Signed-off-by: Xin Yao <xiny@nvidia.com>
Signed-off-by: oliver könig <okoenig@nvidia.com>
Signed-off-by: Cory Ye <cye@nvidia.com>
Signed-off-by: dimapihtar <dpykhtar@nvidia.com>
Signed-off-by: Maanu Grover <maanug@nvidia.com>
Signed-off-by: Oliver Koenig <okoenig@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>
Signed-off-by: Ankur Srivastava <your_verified_email@domain.com>
Signed-off-by: meg miranda <mmiranda@nvidia.com>
Signed-off-by: Youngeun Kwon <youngeunk@nvidia.com>
Co-authored-by: Dennis(Zhenhuan) Liu <denliu@nvidia.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: Philip Petrakian <ppetrakian@nvidia.com>
Co-authored-by: Xin Yao <xiny@nvidia.com>
Co-authored-by: oliver könig <okoenig@nvidia.com>
Co-authored-by: Cory Ye <44509866+cspades@users.noreply.github.com>
Co-authored-by: Dmytro Pykhtar <37850217+dimapihtar@users.noreply.github.com>
Co-authored-by: Maanu Grover <109391026+maanug-nv@users.noreply.github.com>
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
Co-authored-by: Li Tao <lit@nvidia.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: janEbert <janpabloe@nvidia.com>
Co-authored-by: Shifang Xu <shifangx@nvidia.com>
Co-authored-by: Zhiyu Li <zhiyul@NVIDIA.com>
Co-authored-by: melon <49278241+ezioliao@users.noreply.github.com>
Co-authored-by: liaoyang <yliao@siflow.cn>
Co-authored-by: Eric Harper <eharper@nvidia.com>
Co-authored-by: Deepak Narayanan <deepakn94@gmail.com>
Co-authored-by: Teodor-Dumitru Ene <34819528+tdene@users.noreply.github.com>
Co-authored-by: Yuzhong Wang <yuzhongw@nvidia.com>
Co-authored-by: Dhinesh Ponnarasan <160256912+DhineshPonnarasan@users.noreply.github.com>
Co-authored-by: Seonmyeong Bak <sbak@nvidia.com>
Co-authored-by: Charlie Truong <chtruong@nvidia.com>
Co-authored-by: Kunlun Li <94586211+kunlunl@users.noreply.github.com>
Co-authored-by: Kunlun Li <kunlunl@cw-dfw-cs-001-login-02.cm.cluster>
Co-authored-by: wdykas <73254672+wdykas@users.noreply.github.com>
Co-authored-by: William Dykas <wdykas@oci-hsg-cs-001-vscode-03.cm.cluster>
Co-authored-by: root <root@nvl72065-T16.cm.cluster>
Co-authored-by: root <root@nvl72163-T17.cm.cluster>
Co-authored-by: Deepak Narayanan <dnarayanan@nvidia.com>
Co-authored-by: Haoran Zhang <github@snowchord.com>
Co-authored-by: Keshav Santhanam <ksanthanam@nvidia.com>
Co-authored-by: Ankur Srivastava <101727556+awsankur@users.noreply.github.com>
Co-authored-by: Ankur Srivastava <your_verified_email@domain.com>
Co-authored-by: Antoni-Joan Solergibert <asolergibert@nvidia.com>
Co-authored-by: Jon Barker <jbarker@nvidia.com>
Co-authored-by: megnvidia <mmiranda@nvidia.com>
Co-authored-by: Mikail Khona (NVIDIA) <mkhona@nvidia.com>
Co-authored-by: Jingyue Wu <wujingyue@gmail.com>
Co-authored-by: Jorge Albericio <jalbericiola@nvidia.com>
Co-authored-by: Santosh Bhavani <santosh.bhavani@live.com>
Co-authored-by: Youngeun Kwon <youngeunk@nvidia.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
@buptzyb
fix(ci): re-enable scoped_cudagraph MoE test with TE 2.14 golden valu…
8821e6f 
…es (NVIDIA#4444)
@Wohox
[Dev] remove dead manual_release_grads code path in 1F1B overlap sche…
66a2ff8 
…dule (NVIDIA#4510)
@hxbai
[dev] [DeepSeek-v4] Part 2: Hash MoE and SwiGLU clamp (NVIDIA#4481)
fe729e9
@hxbai
[dev] [DeepSeek-v4] Part 1: Hybrid Attention with CSA and HCA (NVIDIA…
bf4e1db 
…#4458)
@shifangx @claude
feat(attention): Add rotary_base_per_layer for Step-3.5-Flash (NVIDIA…
a2d7153 
…#4473)

Co-authored-by: Yu Huang <yuhuang@nvidia.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
@hxbai
MTP support with mHC; new mHC contract
03df3a8
@hxbai
minor fix
da85042
@hxbai
fix new contract dtype
42d79c0
@hxbai
format and add tests
fb32fe2
@yuzhongw-nvidia @hxbai
fix mscale
d9d5eaa
@weijiac0619 @claude
Add hadamard_transform fallback and fix mask dtype for aarch64 container
2f9160d 
- PyTorch fallback for fast_hadamard_transform (unavailable on aarch64/B200)
- Cast mask dtype instead of assert in fused_qk_topk_naive (bf16/fp32 mismatch)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@weijiac0619
hc_head fix
0a55633 
Signed-off-by: weijiac <weijiac@nvidia.com>
@copy-pr-bot
Copy link
Copy Markdown

copy-pr-bot Bot commented May 13, 2026

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.