-
Notifications
You must be signed in to change notification settings - Fork 316
Pull requests: NVIDIA-NeMo/Megatron-Bridge
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[model] fix: Handle Qwen VL MTP with context parallelism
area:model
Model implementations and HF bridge logic
bug
Something isn't working
needs-more-tests
Requires additional L0 and L1 test coverage before merge
#3895
opened May 20, 2026 by
cuichenx
Contributor
Loading…
[tutorial] feat: add MoE notebook example
area:recipe
Training recipes and launch configs
community-request
docs
Documentation-only updates or documentation debt
docs-only
With great power comes great responsibility.
waiting-on-customer
Waiting on the original author to respond
#3890
opened May 19, 2026 by
karinseve
Loading…
1 task
[data] fix: tolerate shared filesystem mkdir races
area:data
Dataset builders, preprocessing, and samplers
bug
Something isn't working
community-request
waiting-on-maintainers
Waiting on maintainers to respond
#3888
opened May 19, 2026 by
janbernloehr
Loading…
4 of 5 tasks
[model] feat: Add limited Gemma4 dense model support
area:model
Model implementations and HF bridge logic
community-request
feature
New capabilities, enhancements, or enablement work
waiting-on-maintainers
Waiting on maintainers to respond
#3885
opened May 19, 2026 by
pavelgein
Contributor
Loading…
2 of 5 tasks
chore(beep boop 🤖): Bump Dependencies, packaging, images, and environment setup
ci
CI, automation, test queue, or workflow infrastructure work
full-test-suite
needs-review
PR is ready for code review and waiting on a reviewer
uv.lock (main, mcore-dev) (2026-05-19)
area:build
#3883
opened May 19, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
chore(beep boop 🤖): Bump Dependencies, packaging, images, and environment setup
ci
CI, automation, test queue, or workflow infrastructure work
full-test-suite
needs-review
PR is ready for code review and waiting on a reviewer
uv.lock (r0.4.1, mcore-core_r0.17.0) (2026-05-19)
area:build
#3882
opened May 19, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
[ckpt, model] refactor: remove unused helper functions
area:ckpt
Checkpoint conversion, loading, export, and save paths
feature
New capabilities, enhancements, or enablement work
needs-review
PR is ready for code review and waiting on a reviewer
#3879
opened May 19, 2026 by
yaoyu-33
Contributor
Loading…
4 of 5 tasks
feat(performance): add --mlperf_flavor for MLPerf v6.0 apples-to-appl…
area:perf
Performance optimizations and benchmarking
feature
New capabilities, enhancements, or enablement work
waiting-on-maintainers
Waiting on maintainers to respond
#3878
opened May 19, 2026 by
rsalagame-nvidia
Contributor
Loading…
[inference] fix: Support MCore dev inference mode
area:model
Model implementations and HF bridge logic
bug
Something isn't working
full-test-suite
needs-review
PR is ready for code review and waiting on a reviewer
#3876
opened May 18, 2026 by
yaoyu-33
Contributor
Loading…
chore: Pin nvidia-cudnn-frontend to 1.23.0
area:build
Dependencies, packaging, images, and environment setup
ci
CI, automation, test queue, or workflow infrastructure work
full-test-suite
waiting-on-customer
Waiting on the original author to respond
#3874
opened May 18, 2026 by
chtruong814
Contributor
Loading…
5 tasks
chore(beep boop 🤖): Bump Dependencies, packaging, images, and environment setup
ci
CI, automation, test queue, or workflow infrastructure work
full-test-suite
needs-review
PR is ready for code review and waiting on a reviewer
uv.lock (r0.4.1, mcore-core_r0.17.0) (2026-05-18)
area:build
#3870
opened May 18, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
[data, model, training] fix: Stabilize Qwen3-VL packed SFT
area:data
Dataset builders, preprocessing, and samplers
area:model
Model implementations and HF bridge logic
area:training
Training loop, callbacks, and runtime integration
bug
Something isn't working
model-qwen
needs-review
PR is ready for code review and waiting on a reviewer
t-seqpacking
#3869
opened May 18, 2026 by
wplf
Contributor
Loading…
4 tasks done
Support for modelopt with MoE QAT
area:quant
Quantization (PTQ, QAT, FP8 recipes)
feature
New capabilities, enhancements, or enablement work
needs-review
PR is ready for code review and waiting on a reviewer
#3866
opened May 17, 2026 by
HollowMan6
Member
Loading…
2 of 5 tasks
chore(beep boop 🤖): Bump Dependencies, packaging, images, and environment setup
ci
CI, automation, test queue, or workflow infrastructure work
full-test-suite
needs-review
PR is ready for code review and waiting on a reviewer
uv.lock (r0.4.1, mcore-core_r0.17.0) (2026-05-17)
area:build
#3863
opened May 17, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
cp: Performance optimizations and benchmarking
cherry-pick
docs
Documentation-only updates or documentation debt
docs-only
With great power comes great responsibility.
needs-review
PR is ready for code review and waiting on a reviewer
r0.4.0
Auto-cherrypick to release branch. Apply before merge; cherrypick happens after merge.
Run CICD
2604_patch_perf_summary (3818) into r0.4.0
area:perf
#3861
opened May 16, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
chore(beep boop 🤖): Bump Dependencies, packaging, images, and environment setup
ci
CI, automation, test queue, or workflow infrastructure work
full-test-suite
needs-review
PR is ready for code review and waiting on a reviewer
uv.lock (r0.4.1, mcore-core_r0.17.0) (2026-05-16)
area:build
#3858
opened May 16, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
[models] refactor: Remove size-specific provider classes
area:model
Model implementations and HF bridge logic
breaking-change
Public behavior or API compatibility changes
feature
New capabilities, enhancements, or enablement work
full-test-suite
waiting-on-maintainers
Waiting on maintainers to respond
#3854
opened May 15, 2026 by
yaoyu-33
Contributor
Loading…
chore(beep boop 🤖): Bump Dependencies, packaging, images, and environment setup
ci
CI, automation, test queue, or workflow infrastructure work
full-test-suite
needs-review
PR is ready for code review and waiting on a reviewer
uv.lock (r0.4.1, mcore-core_r0.17.0) (2026-05-15)
area:build
#3842
opened May 15, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
[training, perf] fix: THD-aware FLOPS via cu_seqlens (Σᵢ sᵢ²)
area:perf
Performance optimizations and benchmarking
bug
Something isn't working
needs-review
PR is ready for code review and waiting on a reviewer
#3839
opened May 15, 2026 by
cuichenx
Contributor
Loading…
[model, perf] feat: real THD packing in qwen3_vl_step
area:model
Model implementations and HF bridge logic
blocked
Work cannot move forward until an external dependency is cleared
feature
New capabilities, enhancements, or enablement work
needs-more-tests
Requires additional L0 and L1 test coverage before merge
#3838
opened May 15, 2026 by
cuichenx
Contributor
Loading…
2 tasks
Add Param2 model bridge
area:model
Model implementations and HF bridge logic
feature
New capabilities, enhancements, or enablement work
needs-more-tests
Requires additional L0 and L1 test coverage before merge
waiting-on-maintainers
Waiting on maintainers to respond
#3834
opened May 14, 2026 by
meghmak13
Loading…
5 tasks
fix(mimo): load optimizer state on resume for MIMO + GLOBAL torch_dist checkpoints
#3832
opened May 14, 2026 by
kamran-nvidia
Contributor
•
Draft
5 tasks
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.