Skip to content

From NVIDIA Megatron-LM for visibility #18

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
RaymondLi0 wants to merge 5,951 commits into
base: multi-query-attention
Choose a base branch
from
Open

From NVIDIA Megatron-LM for visibility #18

RaymondLi0 wants to merge 5,951 commits into from
+555,308 −36,121

Conversation

@RaymondLi0
Copy link
Collaborator

@RaymondLi0 RaymondLi0 commented Jan 24, 2023

No description provided.

@RaymondLi0 RaymondLi0 changed the base branch from multi-query-attention to before-merge June 20, 2023 20:12
@RaymondLi0 RaymondLi0 changed the base branch from before-merge to multi-query-attention June 20, 2023 20:12
chtruong814 and others added 28 commits November 22, 2025 11:10
@chtruong814
ci: Update backwards compat check baseline to 53bbf7a (#2361)
8954e04 
Signed-off-by: Charlie Truong <[email protected]>
@tdene
Account for test regression caused by prints (#2354)
d313c6d
@ananthsub
Remove dependency on megatron.training within megatron.core (#2274)
14464d1 
Signed-off-by: Ananth Subramaniam <[email protected]>
@cuichenx
Fixes for gpt-oss (#2038)
9873958 
Signed-off-by: Chen Cui <[email protected]>
@pablo-garay
update
26b2e72
@Autumn1998
[HOT FIX] Fix bug of hybrid-ep backend in flex-dispatcher (#2286)
326ec8c
@chtruong814
ci: Remove nemo-ci environment (#2364)
17cd106 
Signed-off-by: Charlie Truong <[email protected]>
@chtruong814
ci: Pass COMMUNITY_PROJECT_ID to community bot (#2366)
278e058 
Signed-off-by: Charlie Truong <[email protected]>
@chtruong814
ci: Remove environment from community-bot (#2376)
d61029f 
Signed-off-by: Charlie Truong <[email protected]>
@pablo-garay
monitoring & results in mcore
9269dda
@pablo-garay
Add mbridge_ref input to select MBridge branch
77b65ed
@pablo-garay
Fix: Use correct repo NVIDIA-NeMo/Megatron-Bridge and add mbridge_ref…
aa7a564 
… input
@pablo-garay
gha action
7f70e22 
Signed-off-by: Pablo Garay <[email protected]>
@chtruong814
ci: Bump commit for api check to d61029f (#2386)
c28b84e 
Signed-off-by: Charlie Truong <[email protected]>
@pablo-garay
tidy / synthesize / enhance
ab1e26e 
Signed-off-by: Pablo Garay <[email protected]>
@pablo-garay
Merge branch 'main' of https://github.com/NVIDIA/Megatron-LM
56e8810
@pablo-garay
Revert: trigger_mbridge_tests.yml‎ file change (#2389)
bc242d9
@ko3n1g
build: Upgrade deps (#2289)
49eef58 
Signed-off-by: oliver könig <[email protected]>
@kanz-nv @sidsingh-nvidia
Change KV cache init to empty to speedup graph recording and first pr…
2a51d86 
…efill (#2358)

Co-authored-by: Siddharth Singh <[email protected]>
@tdene
Handle UVM compile lock issues (#2299)
4c7d3d6
@Victarry @yaox12
Remove experimental tags for fused kernels. (#2233)
14b791b 
Co-authored-by: Xin Yao <[email protected]>
@yaox12
Reduce Overhead in Timers (#2210)
ffb8c35 
Signed-off-by: Xin Yao <[email protected]>
@ko3n1g
Revert "build: Upgrade deps (#2289)"
60df5c2 
This reverts commit 49eef58.
@yobibyte @jon-barker @tdene
Fix the entropy sign. (#2374)
ba9caf4 
Co-authored-by: Jon Barker <[email protected]>
Co-authored-by: Teodor-Dumitru Ene <[email protected]>
@jon-barker
Remove RL use of mock dataloader and kill RL inference interface on e…
77a2d8b 
…xit (#2387)

Co-authored-by: Jon Barker <[email protected]>
@kanz-nv @sidsingh-nvidia
Fix block_bag for RL (#2399)
6f65536 
Co-authored-by: Siddharth Singh <[email protected]>
@theothermike
adding action for checking whether PR author is nvidia employee or no…
13efcb8 
…t for selecting ephemeral ci hosts (#2402)
@shanmugamr1992 @ko3n1g
Added top n log probs (#2262)
898d633 
Co-authored-by: Shanmugam Ramasamy <[email protected]>
Co-authored-by: oliver könig <[email protected]>
Co-authored-by: Shanmugam Ramasamy <[email protected]>
ankurv-nvidia and others added 30 commits January 7, 2026 02:50
@ankurv-nvidia @deepakn94
Bugfix: ensure spawned persistent checkpoint worker sets its CUDA dev…
5950971 
…ice correctly for CUDA context creation / hypothetical memory allocations (#2710)

Co-authored-by: Deepak Narayanan <[email protected]>
@jstjohn
Implementation of a more flexible optimizer/scheduler override system (…
41cedd6 
…#2723)

Signed-off-by: John St John <[email protected]>
Signed-off-by: John St. John <[email protected]>
@github-actions
chore: rotate oncall schedule
64bc482
@ko3n1g
ci(fix): PyPI upload (#2843)
144049d 
Signed-off-by: oliver könig <[email protected]>
@ko3n1g
ci(fix): Don't fail on empty var (#2850)
ed5bc5c 
Signed-off-by: oliver könig <[email protected]>
@jon-barker
Add RL support for MOEs (#2742)
c64f227
@ko3n1g
ci(fix): GH release version tag (#2854)
ae1a2f1 
Signed-off-by: oliver könig <[email protected]>
@ko3n1g
ci(hotfix): Disable flaky test
7e5e16b 
Signed-off-by: oliver könig <[email protected]>
@jstjohn
Reduce the scope of the side stream around DDP initialization (#2852)
4ba9f46 
Signed-off-by: John St. John <[email protected]>
@Phlip79
Manually update first oncall rotation (#2855)
49dfee2
@buptzyb
Remove flaky iteration time functional test (#2862)
5fa42ec 
Signed-off-by: Robin Zhang <[email protected]>
@wdykas
Nccl gloo refit for RL (#2812)
0cc98c4 
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: William Dykas <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
Co-authored-by: root <[email protected]>
@ko3n1g
build: Bump jet-client (#2876)
8d6c604 
Signed-off-by: oliver könig <[email protected]>
@Phlip79
Change oncall team name (#2861)
8b10a64
@lmcafee-nvidia
Dynamic Inference | Evict and re-compute context requests. (#2738)
c8ac1fe
@buptzyb
Fix CUDA RNG Tracker (#2641)
965cfd3 
Signed-off-by: Robin Zhang <[email protected]>
@yuzhongw-nvidia
[main] feat(moe): Support attention output gate for Qwen3-Next (3/4) (#…
65dccab 
…2752)
@yuzhongw-nvidia @Phlip79
[main] feat(moe): Support moe shared expert gate for Qwen3-Next (2/4) (…
bddd0a8 
…#2751)

Co-authored-by: Philip Petrakian <[email protected]>
@Phlip79
[docs] Fix docs and add generation doc (#2882)
0b5be24
@chtruong814
Revert "Dynamic Inference | Evict and re-compute context requests. (#…
43b4471 
…2738)" (#2884)

Signed-off-by: Charlie Truong <[email protected]>
@kunlunl @shjwudp @cspades
FP8 params support for megatron-fsdp (MXFP8/Blockwise) (#2239)
8f2f700 
Signed-off-by: kunlunl <[email protected]>
Signed-off-by: jianbinc <[email protected]>
Co-authored-by: jianbinc <[email protected]>
Co-authored-by: Cory Ye <[email protected]>
@sbhavani @yaox12
docs: fix broken images, links, and typos across documentation (#2794)
ed29157 
Co-authored-by: Xin Yao <[email protected]>
@ko3n1g
ci(fix): Release version (#2873)
424a26d 
Signed-off-by: oliver könig <[email protected]>
@Phlip79
Assign mcore-oncall instead of user (#2879)
371ee52
@ko3n1g
tests: Disable Mamba MOE model test after 43b4471 (#2886)
74db1ce 
Signed-off-by: oliver könig <[email protected]>
@jon-barker
Fix mamba moe unit test after commit reversion (#2888)
7fe7f48
@jon-barker
Improve error messages in mamba moe unit test (#2889)
ed461d6
@santhnm2 @shanmugamr1992
Use DynamicInferenceCoordinator for text generation server (#1910)
c0b2859 
Signed-off-by: Keshav Santhanam <[email protected]>
Co-authored-by: shanmugamr1992 <[email protected]>
Co-authored-by: Shanmugam Ramasamy <[email protected]>
@yobibyte
Fix inference server to make nemogym work. (#2887)
980a271
@maanug-nv
[training migration] add RNG config dataclass (#2347)
4e0eac4 
Signed-off-by: Maanu Grover <[email protected]>
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.