Skip to content

test: gb200 nightlies/release tests#1766

Closed
terrykong wants to merge 19 commits intor0.5.0from
chtruong/r0.5.0-gb200
Closed

test: gb200 nightlies/release tests#1766
terrykong wants to merge 19 commits intor0.5.0from
chtruong/r0.5.0-gb200

Conversation

@terrykong
Copy link
Copy Markdown
Collaborator

@terrykong terrykong commented Jan 13, 2026 •
edited
Loading

Needs to be cherry-picked against main when ready so we can cherry-pick back into r0.5

terrykong and others added 17 commits January 11, 2026 16:23
@terrykong @chtruong814
tests: gb200 nightlies/release
d02b0bc 
Signed-off-by: Terry Kong <terryk@nvidia.com>

run config cli test first b/c other run_first tests depend on its
correctness

Signed-off-by: Terry Kong <terryk@nvidia.com>
@terrykong @chtruong814
5 steps
393d32c 
Signed-off-by: Terry Kong <terryk@nvidia.com>
@chtruong814
Revert "5 steps"
883a0aa 
This reverts commit 8ca7edc.
@chtruong814
Update gb200 test times
d16ef81 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Allow excluding gres arg in tools/launch
836a124 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Fix gres arg exclusion in launch script
1b213e9 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Update steps per run for nano-v2 for testing launch
d7648ff 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Bump mamba to d68d16e and causal-conv1d to 67e0a9d
81c3e4c 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Increase gb200 nanov3 testing
e40d2cf 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Fix uv lock
b00ffbc 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Update grpo-nano-v2-12b-1n4g-megatron to run in 3 steps
74d69b9 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Fix gb200 configs
e22b025 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Revert "Update grpo-nano-v2-12b-1n4g-megatron to run in 3 steps"
4097efd 
This reverts commit 74d69b9.

Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Revert "Update steps per run for nano-v2 for testing launch"
46f7d9d 
This reverts commit d7648ff.

Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Fix setting colocated gpus per node for llam3.1 8b test
9875f13 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Update time config for deepscalar and llama3.2 gb200 tests
9995e63 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Merge remote-tracking branch 'origin/r0.5.0' into chtruong/r0.5.0-gb200
aa791db 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@terrykong terrykong requested review from a team as code owners January 13, 2026 07:00
@github-actions github-actions Bot added the Documentation Improvements or additions to documentation label Jan 13, 2026
@terrykong terrykong mentioned this pull request Jan 13, 2026
Closed
@terrykong terrykong changed the base branch from main to r0.5.0 January 13, 2026 07:01
@terrykong terrykong marked this pull request as draft January 13, 2026 07:01
@terrykong terrykong mentioned this pull request Jan 13, 2026
Merged
@chtruong814
Adjust release distillation steps per prun
a2b1334 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@chtruong814
Pass model name to DSv3 release test config
82566bf 
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
@terrykong
Copy link
Copy Markdown
Collaborator Author

terrykong commented Jan 23, 2026

merged in #1789

@terrykong terrykong closed this Jan 23, 2026
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

Documentation Improvements or additions to documentation

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@terrykong @chtruong814