Fetch from nvidia Megatron-LM by RaymondLi0 · Pull Request #5 · ElementAI/Megatron-LM

RaymondLi0 · 2022-08-03T20:32:25Z

No description provided.

Signed-off-by: Cory Ye <cye@nvidia.com> Co-authored-by: Cory Ye <cye@nvidia.com> Co-authored-by: conver334 <conver334@gmail.com>

Signed-off-by: Faradawn Yang <73060648+faradawn@users.noreply.github.com>

Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>

…graphs) (#4085) Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

… workflow (#4199) Signed-off-by: oliver könig <okoenig@nvidia.com> Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

…ment-wise distributed optimizer (#4138) Signed-off-by: Hao Wu <skyw@nvidia.com>

Signed-off-by: oliver könig <okoenig@nvidia.com> Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

Signed-off-by: Charlie Truong <chtruong@nvidia.com>

Signed-off-by: Hao Wu <skyw@nvidia.com>

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

Signed-off-by: Maanu Grover <maanug@nvidia.com>

Signed-off-by: oliver könig <okoenig@nvidia.com> Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

Signed-off-by: xiaoxi-wangfj <690912414@qq.com> Co-authored-by: Xin Yao <xiny@nvidia.com>

Co-authored-by: eternally-z <zzywzj@gmail.com> Co-authored-by: Philip Petrakian <ppetrakian@nvidia.com>

Co-authored-by: gautham-kollu <gkollu@nvidia.com>

) Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Signed-off-by: jeffnvidia <jmahou@nvidia.com> Co-authored-by: Cory Ye <44509866+cspades@users.noreply.github.com>

Co-authored-by: gautham-kollu <gkollu@nvidia.com> Co-authored-by: Xin Yao <xiny@nvidia.com>

Co-authored-by: Antoni-Joan Solergibert <asolergibert@nvidia.com> Co-authored-by: gautham-kollu <gkollu@nvidia.com> Co-authored-by: Xin Yao <xiny@nvidia.com>

…n_dev (#4639) Signed-off-by: oliver könig <okoenig@nvidia.com>

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

…expert overlap in latent MoEs (#4603)

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

Signed-off-by: Charlie Truong <chtruong@nvidia.com>

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

Signed-off-by: Charlie Truong <chtruong@nvidia.com>

Signed-off-by: oliver könig <okoenig@nvidia.com>

…q_len (#4094)

…#4712) Signed-off-by: oliver könig <okoenig@nvidia.com>

Signed-off-by: oliver könig <okoenig@nvidia.com>

Co-authored-by: Antoni-Joan Solergibert <asolergibert@nvidia.com>

…ename seq_len (#4094)" (#4718) Signed-off-by: oliver könig <okoenig@nvidia.com>

…al tests` (#4730) Signed-off-by: oliver könig <okoenig@nvidia.com>

Signed-off-by: oliver könig <okoenig@nvidia.com>

Signed-off-by: oliver könig <okoenig@nvidia.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Signed-off-by: oliver könig <okoenig@nvidia.com>

Signed-off-by: Shivanjan Chakravorty <shivanjanc@nvidia.com> Co-authored-by: Cory Ye <44509866+cspades@users.noreply.github.com>

ko3n1g force-pushed the main branch from ad5a627 to cd71d4c Compare February 18, 2026 09:18

shjwudp and others added 29 commits April 7, 2026 18:51

[M-FSDP] Fix Tensor Parallel mode detection (#3191)

8cbc45b

Signed-off-by: Cory Ye <cye@nvidia.com> Co-authored-by: Cory Ye <cye@nvidia.com> Co-authored-by: conver334 <conver334@gmail.com>

fix: remove weights_only=False for multimodal example (#4104)

8e8aff6

Signed-off-by: Faradawn Yang <73060648+faradawn@users.noreply.github.com>

Cudagraphs: Fix sequence packing segfault more generally (#4162)

52150ab

Update copy-pr-bot.yaml [skip ci]

70a7f69

Make MTP work with materialize_only_last_token_logits (#4166)

c146305

Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>

Add unit test for Mamba EP inference (eager fallback with mixed CUDA …

8cf6b35

…graphs) (#4085) Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>

chore: rotate oncall schedule

2368a2e

update docs in respect to async changes (#4177)

cc025f8

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

update checkpointing docs in respect to async changes (#4208)

1341d8c

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

chore: improve build-and-test skill with trigger rules and dependency…

e40feed

… workflow (#4199) Signed-off-by: oliver könig <okoenig@nvidia.com> Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

Fix layerwise optimizer with expt_dp_size=1 and contention with ele…

51bcf14

…ment-wise distributed optimizer (#4138) Signed-off-by: Hao Wu <skyw@nvidia.com>

ci: add --cluster-a100/h100/gb200 args to trigger_internal_ci.py (#4195)

50bafa0

Signed-off-by: oliver könig <okoenig@nvidia.com> Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

ci: Update golden values for nightly tests (#4215)

41595d0

Signed-off-by: Charlie Truong <chtruong@nvidia.com>

Update copy-pr-bot.yaml [skip ci]

a614df4

rename async_allgather to overlap_param_gather (#4217)

1b425ab

Signed-off-by: Hao Wu <skyw@nvidia.com>

Fix Slack sync for users with GitHub email privacy enabled (#4220)

8f72c54

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Miscellaneous MTP inference fixes (#4191)

980211a

Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>

Move inference guards out of arguments.py (#4210)

be09bb6

Fix: enable fine-grained activation offloading for Mamba model. (#4173)

92d5c1f

bump NVRx (#4178)

b5354a8

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

Update tokenizer args for Nemotron3 release config (#4239)

09312c8

Signed-off-by: Maanu Grover <maanug@nvidia.com>

build: add dynamic git-versioning and drop rc0 pre-release tag (#4212)

2f5c62c

Signed-off-by: oliver könig <okoenig@nvidia.com> Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

Fix unnecessary permute padding for non-quantized MoE dispatch (#4038)

567d4d4

Signed-off-by: xiaoxi-wangfj <690912414@qq.com> Co-authored-by: Xin Yao <xiny@nvidia.com>

Fix split state dict main (#3676)

51b0950

Co-authored-by: eternally-z <zzywzj@gmail.com> Co-authored-by: Philip Petrakian <ppetrakian@nvidia.com>

Enable FP8 DPA for MXFP8 recipe (#4066)

22e0bb5

Co-authored-by: gautham-kollu <gkollu@nvidia.com>

Add /split-pr Claude Code command for splitting PRs by CODEOWNERS (#4160

3d4de97

) Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Enable AG/RS overlap with explicit process group passing (#3249)

2ebfbb2

Signed-off-by: jeffnvidia <jmahou@nvidia.com> Co-authored-by: Cory Ye <44509866+cspades@users.noreply.github.com>

Enable cpu_offloading with Full iteration CUDA graph (#3969)

d30c3ae

Co-authored-by: gautham-kollu <gkollu@nvidia.com> Co-authored-by: Xin Yao <xiny@nvidia.com>

Fix TransformerConfig validation for mixed dense/MoE upcycling (#3647)

e8e79a4

Co-authored-by: Antoni-Joan Solergibert <asolergibert@nvidia.com> Co-authored-by: gautham-kollu <gkollu@nvidia.com> Co-authored-by: Xin Yao <xiny@nvidia.com>

ko3n1g and others added 30 commits May 6, 2026 10:14

test: mark TestFusedApplyMLARope::test_forward_backward_for_q flaky_i…

c325855

…n_dev (#4639) Signed-off-by: oliver könig <okoenig@nvidia.com>

chore: rotate oncall schedule

fd443f2

remove legacy GPT code (#4322)

39ec5eb

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

ci: introduce L-tier scope vocabulary via parser (#4625)

008c3e9

Inference: Tune vLLM grouped gemm, moe_sum kernel, and enable shared …

20f0936

…expert overlap in latent MoEs (#4603)

Fix crash involving evicted requests and tpot (#4645)

5c7968a

remove legacy tranformer and modules (#4207)

8de8238

Signed-off-by: dimapihtar <dpykhtar@nvidia.com>

chore: Update Docker image version to 26.04-py3 (#4611)

5fe3f06

Propagate errors for failed inference requests (#4679)

387ba42

Inference: Cache input + position ID views (#4634)

1df264c

ci: Update Gitlab base image to 26.04 pytorch (#4688)

7cdf652

Signed-off-by: Charlie Truong <chtruong@nvidia.com>

Add periodic GPU sniff tests to detect hardware stragglers (#4662)

932d9ee

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

ci: Bump GHA versions (#4606)

8079f37

Signed-off-by: Charlie Truong <chtruong@nvidia.com>

build: widen flashinfer-python pin to <0.7.0 (#4700)

a7bd379

Signed-off-by: oliver könig <okoenig@nvidia.com>

Add Python-side guardrail for HybridEP InfiniBand limit and rename se…

a08e259

…q_len (#4094)

Switch oncall (#4702)

dfd4a1a

Update golden values for various functional tests (#4703)

4933b24

chore: Update golden values for various functional tests (#4706)

31fbdf0

build: upgrade mamba-ssm to 2.3.2.post1, causal-conv1d to 1.6.2.post1 (…

708c6f5

…#4712) Signed-off-by: oliver könig <okoenig@nvidia.com>

ci: replace uuidgen with /proc/sys/kernel/random/uuid (#4714)

1256299

Signed-off-by: oliver könig <okoenig@nvidia.com>

chore(codeowners): add megatron/inference/ ownership (#4704)

7106fe9

Signed-off-by: oliver könig <okoenig@nvidia.com>

Create a Protocol for the MLP layer of TransformerLayer (#3435)

5e31514

Co-authored-by: Antoni-Joan Solergibert <asolergibert@nvidia.com>

Revert "Add Python-side guardrail for HybridEP InfiniBand limit and r…

a2ec5c1

…ename seq_len (#4094)" (#4718) Signed-off-by: oliver könig <okoenig@nvidia.com>

chore(beep boop 🤖): Bump (main) (2026-05-11)

e93755e

Add Python-side guardrail for DeepEP IB limits (#4719)

ad58411

ci: revert bad uv.lock bump and label future bumps with `Run function…

5123f6a

…al tests` (#4730) Signed-off-by: oliver könig <okoenig@nvidia.com>

[ci] fix: treat cancelled run-main-script step as failure (#4727)

33d47e0

Signed-off-by: oliver könig <okoenig@nvidia.com>

ci: Major refactor of release-workflows (#4602)

e42e2fa

Signed-off-by: oliver könig <okoenig@nvidia.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

build(deps): bump nvidia-modelopt to 0.43 (#4723)

434368c

Signed-off-by: oliver könig <okoenig@nvidia.com>

fix(fsdp): recognize legacy GDN TP metadata (#4664)

74687fe

Signed-off-by: Shivanjan Chakravorty <shivanjanc@nvidia.com> Co-authored-by: Cory Ye <44509866+cspades@users.noreply.github.com>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fetch from nvidia Megatron-LM#5

Fetch from nvidia Megatron-LM#5
RaymondLi0 wants to merge 7255 commits into
ElementAI:load-iterfrom
NVIDIA:main

RaymondLi0 commented Aug 3, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

Conversation

RaymondLi0 commented Aug 3, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants