Skip to content

[WIP] Enable Dynamic Context Parallelism for long context training#2133

Draft
parthmannan wants to merge 23 commits intoNVIDIA-NeMo:mainfrom
parthmannan:pmannan/long_context_rl_perf
Draft

[WIP] Enable Dynamic Context Parallelism for long context training#2133
parthmannan wants to merge 23 commits intoNVIDIA-NeMo:mainfrom
parthmannan:pmannan/long_context_rl_perf

Conversation

@parthmannan
Copy link
Contributor

@parthmannan parthmannan commented Mar 20, 2026

What does this PR do ?

This work is still WIP and not yet ready for review.

Issues

List issues that this PR closes (syntax):

Usage

  • You can potentially add a usage example below
# Add a code snippet demonstrating how to use this

Before your PR is "Ready for review"

Pre checks:

  • Make sure you read and followed Contributor guidelines
  • Did you write any new necessary tests?
  • Did you run the unit tests and functional tests locally? Visit our Testing Guide for how to run tests
  • Did you add or update any necessary documentation? Visit our Document Development Guide for how to write, build and test the docs.

Additional Information

  • ...

guyueh1 and others added 21 commits January 20, 2026 08:30
Signed-off-by: Guyue Huang <guyueh@nvidia.com>
Signed-off-by: Guyue Huang <guyueh@nvidia.com>
Signed-off-by: Guyue Huang <guyueh@nvidia.com>
…com/guyueh1/NeMo-RL into time_data_sharding_and_job_submission
Signed-off-by: Guyue Huang <guyueh@nvidia.com>
Signed-off-by: Guyue Huang <guyueh@nvidia.com>
Signed-off-by: Yi-Fu Wu <yifu.wu@gmail.com>
Signed-off-by: Yi-Fu Wu <yifu.wu@gmail.com>
Signed-off-by: Yi-Fu Wu <yifu.wu@gmail.com>
Signed-off-by: Yi-Fu Wu <yifu.wu@gmail.com>
…/long_context_rl_perf

# Conflicts:
#	.gitmodules
#	3rdparty/Megatron-Bridge-workspace/setup.py
#	3rdparty/Megatron-LM-workspace/Megatron-LM
#	3rdparty/Megatron-LM-workspace/setup.py
#	nemo_rl/models/megatron/common.py
#	nemo_rl/models/megatron/setup.py
#	nemo_rl/models/policy/lm_policy.py
#	nemo_rl/models/policy/workers/megatron_policy_worker.py
#	uv.lock
@copy-pr-bot
Copy link

copy-pr-bot bot commented Mar 20, 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.

@github-actions
Copy link

✅ Submodule Fast-Forward Check Results

Check based on commit: 6ae418b (PR #2133 from pmannan/long_context_rl_perf)

✅ Submodules that are properly updated:

Megatron-LM: ✅ PR branch is ahead of main branch (fast-forward)

All submodule changes look good! ✨

@github-actions
Copy link

✅ Submodule Fast-Forward Check Results

Check based on commit: c9cc885 (PR #2133 from pmannan/long_context_rl_perf)

✅ Submodules that are properly updated:

Megatron-LM: ✅ PR branch is ahead of main branch (fast-forward)

All submodule changes look good! ✨

Signed-off-by: Parth Mannan <pmannan@nvidia.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants