Skip to content

fix: support Qwen-Image finetune (T2I)#1704

Open
harshareddy832 wants to merge 10 commits intoNVIDIA-NeMo:mainfrom
harshareddy832:fix/1700-support-qwen-image-finetune-t2i
Open

fix: support Qwen-Image finetune (T2I)#1704
harshareddy832 wants to merge 10 commits intoNVIDIA-NeMo:mainfrom
harshareddy832:fix/1700-support-qwen-image-finetune-t2i

Conversation

@harshareddy832
Copy link
Copy Markdown
Contributor

@harshareddy832 harshareddy832 commented Apr 7, 2026

Fixes #1700

Changes:

@copy-pr-bot
Copy link
Copy Markdown

copy-pr-bot bot commented Apr 7, 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.

@harshareddy832 harshareddy832 changed the title Support Qwen-Image finetune (T2I) fix: support Qwen-Image finetune (T2I) Apr 7, 2026
akoumpa
akoumpa reviewed Apr 7, 2026
View reviewed changes
Comment thread .claude/settings.json Outdated
Copy link
Copy Markdown
Contributor

@akoumpa akoumpa Apr 7, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hi @harshareddy832 thanks for making this PR, would it be ok to remove this file to avoid any conflicts with other people's files (i know it's project but..)? :thanks:

@akoumpa
Copy link
Copy Markdown
Contributor

akoumpa commented Apr 7, 2026

Thanks a lot @harshareddy832 , have you tested this by any chance?

@akoumpa
Copy link
Copy Markdown
Contributor

akoumpa commented Apr 7, 2026

/ok to test 27ee775

@pthombre
Copy link
Copy Markdown
Contributor

pthombre commented Apr 7, 2026

Hey @harshareddy832 . Thanks for this contribution. Could you please share wandb logs from a recent training run? It would be great to include this information in the PR.

@harshareddy832
Copy link
Copy Markdown
Contributor Author

harshareddy832 commented Apr 7, 2026

Hey @akoumpa and @pthombre, thanks for the feedback!

@akoumpa — good catch on .claude/settings.json, I've removed it. And yes, I've done some basic testing locally — the flow matching tests pass cleanly.

@pthombre — I'll get a training run going and share the wandb logs shortly. Haven't had a chance to run a full training experiment yet but will update the PR once I do.

Also cleaned up a small import style nit while I was at it (multi-line import for QwenImageAdapter).

@pthombre
Copy link
Copy Markdown
Contributor

pthombre commented Apr 7, 2026

/claude review

claude[bot]
claude bot reviewed Apr 7, 2026
View reviewed changes
Comment thread nemo_automodel/components/flow_matching/adapters/qwen_image.py
harshareddy832 and others added 6 commits April 7, 2026 23:25
@harshareddy832
fix: support qwen-image finetune (t2i) (NVIDIA-NeMo#1700)
0ec2117 
Signed-off-by: Harsha Pasham <pashamharsha018@gmail.com>
@harshareddy832
Delete .claude/settings.json
6bf7378 
Signed-off-by: Harsha Pasham <pashamharsha018@gmail.com>

Signed-off-by: Harsha Pasham <pashamharsha018@gmail.com>
@harshareddy832
fix: support qwen-image finetune (t2i) (NVIDIA-NeMo#1700)
72c591f 
Signed-off-by: Harsha Pasham <pashamharsha018@gmail.com>
@harshareddy832
chore: remove .claude/settings.json to avoid conflicts
9db8281 
Remove personal editor config file from the repository as it can
conflict with other contributors' settings.

Signed-off-by: Harsha Pasham <pashamharsha018@gmail.com>
@harshareddy832
fix: support qwen-image finetune (t2i) (NVIDIA-NeMo#1700)
826cf71 
Signed-off-by: Harsha Pasham <pashamharsha018@gmail.com>
@harshareddy832
fix: support qwen-image finetune (t2i) (NVIDIA-NeMo#1700)
12dfc5c 
Signed-off-by: Harsha Pasham <pashamharsha018@gmail.com>
@harshareddy832 harshareddy832 force-pushed the fix/1700-support-qwen-image-finetune-t2i branch from be7887c to 12dfc5c Compare April 8, 2026 03:25
@chtruong814 chtruong814 added waiting-for-customer Waiting for response from the original author needs-follow-up Issue needs follow-up and removed waiting-for-customer Waiting for response from the original author labels Apr 14, 2026
@akoumpa
Copy link
Copy Markdown
Contributor

akoumpa commented Apr 19, 2026

/ok to test db3cdbb

@akoumpa
Copy link
Copy Markdown
Contributor

akoumpa commented Apr 19, 2026

/ok to test 74d6f89

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@akoumpa akoumpa akoumpa left review comments

@claude claude[bot] claude[bot] left review comments

@HuiyingLi HuiyingLi Awaiting requested review from HuiyingLi HuiyingLi is a code owner

@adil-a adil-a Awaiting requested review from adil-a adil-a is a code owner

@hemildesai hemildesai Awaiting requested review from hemildesai hemildesai is a code owner

@ZhiyuLi-Nvidia ZhiyuLi-Nvidia Awaiting requested review from ZhiyuLi-Nvidia ZhiyuLi-Nvidia is a code owner

@pthombre pthombre Awaiting requested review from pthombre pthombre is a code owner

At least 1 approving review is required to merge this pull request.

Assignees

No one assigned

Labels

community-request needs-follow-up Issue needs follow-up

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Support Qwen-Image finetune (T2I)

5 participants

@harshareddy832 @akoumpa @pthombre @chtruong814 @svcnvidia-nemo-ci