Skip to content

kimi 2.5#2275

Open
FDecaYed wants to merge 1 commit intoNVIDIA-NeMo:kimi25vlfrom
FDecaYed:deyuf/kimi25vl
Open

kimi 2.5#2275
FDecaYed wants to merge 1 commit intoNVIDIA-NeMo:kimi25vlfrom
FDecaYed:deyuf/kimi25vl

Conversation

@FDecaYed
Copy link
Contributor

@FDecaYed FDecaYed commented Feb 9, 2026

…ct forward pass

  • Rewrite kimi_k25_vl_provider.py (was a duplicate of bridge file) as a proper KimiK25VLModelProvider inheriting from KimiK2Provider
  • Fix aux_loss_alpha and media_placeholder_token_id read from wrong config
  • Fix token ID defaults to match Kimi K2.5 (not Qwen)
  • Fix forward pass: remove breakpoint(), skip transpose when no images, remove _compute_attention_mask (wrong shape after sequence expansion)
  • Add KimiK2Provider to models/init.py, remove unused imports

What does this PR do ?

Add a one line overview of what this PR aims to accomplish.

Changelog

  • Add specific line by line info of high level changes in this PR.

GitHub Actions CI

See the CI sectionin the Contributing doc for how to trigger the CI. A Nvidia developer will need to approve and trigger the CI for external contributors.

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 add or update any necessary documentation?
  • Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
    • Reviewer: Does the PR have correct import guards for all optional libraries?

If you haven't finished some of the above items you can still open "Draft" PR.

Additional Information

  • Related to # (issue)

…ct forward pass

- Rewrite kimi_k25_vl_provider.py (was a duplicate of bridge file) as a
  proper KimiK25VLModelProvider inheriting from KimiK2Provider
- Fix aux_loss_alpha and media_placeholder_token_id read from wrong config
- Fix token ID defaults to match Kimi K2.5 (not Qwen)
- Fix forward pass: remove breakpoint(), skip transpose when no images,
  remove _compute_attention_mask (wrong shape after sequence expansion)
- Add KimiK2Provider to models/__init__.py, remove unused imports

Signed-off-by: Deyu Fu <deyuf@nvidia.com>
@copy-pr-bot
Copy link

copy-pr-bot bot commented Feb 9, 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.

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.

1 participant