Skip to content

Monkey patch deepseek-ocr's v_head_dim#15384

Merged
hnyls2002 merged 2 commits intomainfrom
lsyin/fix-ocr
Dec 18, 2025
Merged

Monkey patch deepseek-ocr's v_head_dim#15384
hnyls2002 merged 2 commits intomainfrom
lsyin/fix-ocr

Conversation

@hnyls2002
Copy link
Copy Markdown
Collaborator

@hnyls2002 hnyls2002 commented Dec 18, 2025

I do not know why https://huggingface.co/deepseek-ai/DeepSeek-OCR/blob/main/config.json#L116.

This will lead to wrong KV cache cell size calculation after #15207.

@gemini-code-assist
Copy link
Copy Markdown
Contributor

gemini-code-assist Bot commented Dec 18, 2025

Summary of Changes

Hello @hnyls2002, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses a critical configuration anomaly in DeepSeek-OCR models where the v_head_dim is erroneously set to zero. By implementing a targeted override during configuration loading, it prevents potential runtime issues related to KV cache management, thereby ensuring the stable and correct operation of these models within the system.

Highlights

  • DeepSeek-OCR v_head_dim Correction: This PR introduces a fix for DeepSeek-OCR models where the v_head_dim parameter is incorrectly set to 0 in their Hugging Face configuration, which could lead to issues with KV cache cell size calculation after a recent update.
  • Conditional Configuration Override: A new utility function, _override_deepseek_ocr_v_head_dim, has been added to detect DeepSeek-OCR models with a v_head_dim of 0 and programmatically override it to 128, along with a warning log.
  • Integration into Model Configuration Loading: The _override_deepseek_ocr_v_head_dim function is now invoked within the get_config utility, ensuring that the v_head_dim correction is applied automatically whenever a DeepSeek-OCR model's configuration is loaded.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist[bot]
gemini-code-assist Bot reviewed Dec 18, 2025
View reviewed changes
Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

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

Code Review

This pull request introduces a monkey patch to address an incorrect v_head_dim value in the deepseek-ocr model's configuration. The change correctly identifies the problematic configuration and overrides it to prevent issues with KV cache size calculations. The implementation is straightforward and well-placed within the get_config function. My feedback focuses on a minor code style improvement to enhance maintainability by replacing a magic number with a named constant.

Comment thread python/sglang/srt/utils/hf_transformers_utils.py Outdated
@hnyls2002 @gemini-code-assist
Update python/sglang/srt/utils/hf_transformers_utils.py
4e6a0f6 
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
@hnyls2002
Copy link
Copy Markdown
Collaborator Author

hnyls2002 commented Dec 18, 2025

/tag-and-rerun-ci

@hnyls2002 hnyls2002 merged commit 8b0a68f into main Dec 18, 2025
77 of 94 checks passed
@hnyls2002 hnyls2002 deleted the lsyin/fix-ocr branch December 18, 2025 08:28
Liwansi added a commit to iforgetmyname/sglang that referenced this pull request Dec 19, 2025
@Liwansi
Merge branch 'main' of https://github.com/sgl-project/sglang into qwe…
b5a509d 
…n3_pp

* 'main' of https://github.com/sgl-project/sglang: (74 commits)
  [bug fix][pp] fix inconsistent latency between tp (sgl-project#15379)
  Fix warp illegal instruction in kimi k2 thinking PCG (sgl-project#15306)
  Fix gpt-oss yarn with `truncate` argument (sgl-project#14270)
  Monkey patch deepseek-ocr's `v_head_dim` (sgl-project#15384)
  [model-gateway] Replace PolicyRegistry RwLock with DashMap for lock-free policy lookups (sgl-project#15361)
  [PP] Fix dynamic chunking strategy for PP (sgl-project#15372)
  Fix issue: ENABLE_BELOW_SM90 cannot be enabled on aarch64 CPU (sgl-project#12967)
  Split test_piecewise_cuda_graph.py to optimize CI resource usage (sgl-project#15290)
  unified management of environment variables for vlm cuda ipc transport  (sgl-project#14501)
  Mistral Large 3 NVFP4 TRTLLM MoE support (sgl-project#15049)
  fix: adjust time for test_epd_disaggregation.py (sgl-project#15354)
  Add doc for qwen3 next (sgl-project#15337)
  feat: DeepSeek-V3.2 Streaming tool call output (sgl-project#15278)
  Feature/trtllm mha workspace size configurable sgl-project#15089 (sgl-project#15131)
  [VLM] Support cos sin cache for Qwen3-VL & GLM-4.1V (sgl-project#15205)
  [Deepseek V3.2] Support Overlap Spec + NSA (sgl-project#15307)
  Add request-level timestamp for when prefill finishes (sgl-project#14860)
  [CI] Migrate LoRA tests to test/registered/lora/ (sgl-project#15176)
  Reserve more memory for DeepSeekOCR model and adjust server start timeout for DeepGEMM to reduce flakiness (sgl-project#15277)
  Fix condition check for require_gathered_buffer (sgl-project#15328)
  ...
Prozac614 pushed a commit to Prozac614/sglang that referenced this pull request Dec 23, 2025
@hnyls2002 @gemini-code-assist @Prozac614
Monkey patch deepseek-ocr's v_head_dim (sgl-project#15384)
8a41b35 
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
jiaming1130 pushed a commit to zhuyijie88/sglang that referenced this pull request Dec 25, 2025
@hnyls2002 @gemini-code-assist @jiaming1130
Monkey patch deepseek-ocr's v_head_dim (sgl-project#15384)
ee3a0e6 
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
YChange01 pushed a commit to YChange01/sglang that referenced this pull request Jan 13, 2026
@hnyls2002 @gemini-code-assist @YChange01
Monkey patch deepseek-ocr's v_head_dim (sgl-project#15384)
7b99e15 
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@yhyang201 yhyang201 yhyang201 approved these changes

+1 more reviewer

@gemini-code-assist gemini-code-assist[bot] gemini-code-assist[bot] left review comments

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

run-ci

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@hnyls2002 @yhyang201