2/5 support kimi 2.5 full + lora: VL-aware quantization/conversion tools - #1220
Merged
yueming-yuan merged 3 commits intoMay 28, 2026
Merged
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request updates quantization and conversion tools to exclude vision-related modules (vision_tower and mm_projector) from quantization, and adapts the Kimi INT4 to BF16 conversion script to support Kimi-K2.5 by handling nested text_config and copying additional file types (.jinja and .model). The review feedback recommends assigning dummy values to unused configuration variables in convert_kimi_int4_to_bf16.py to prevent potential TypeError exceptions, and suggests refactoring the file extension checks to improve readability.
nanjiangwill
changed the title
nanjiangwill
force-pushed
the
kimi25-pr3-vl-convert-tools
branch
from
May 26, 2026 23:53
ab3aa90 to
0503c16
Compare
yueming-yuan
approved these changes
May 28, 2026
yueming-yuan
left a comment
yueming-yuan
left a comment
Collaborator
There was a problem hiding this comment.
LGTM, please check Gemini comment
…tool INT4/FP8 quantization tools now leave vision_tower and mm_projector weights unquantized (VL components stay in their native precision). Rename convert_k2_thinking_int4_to_bf16.py to convert_kimi_int4_to_bf16.py and generalize it for the Kimi K2.5 checkpoints. The original Kimi K2.5 work this is split from is by JiLi (@GeLee-Q). Co-authored-by: JiLi <22428217+GeLee-Q@users.noreply.github.com>
nanjiangwill
force-pushed
the
kimi25-pr3-vl-convert-tools
branch
from
May 28, 2026 04:31
0503c16 to
02cb2c7
Compare
Frankstein73
pushed a commit
to sii-avalanche/miles
that referenced
this pull request
Jun 1, 2026
…ols (radixark#1220) Co-authored-by: JiLi <22428217+GeLee-Q@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Part 2/5 of splitting #1057 (Kimi K2.5 full-param + LoRA support) into reviewable PRs, rebased on latest main.
INT4/FP8 quantization tools leave
vision_towerandmm_projectorweights unquantized (VL components stay in native precision). Renameconvert_k2_thinking_int4_to_bf16.pytoconvert_kimi_int4_to_bf16.pyand generalize it for Kimi K2.5 checkpoints.Independent of the other PRs (disjoint files).