Skip to content

feat: add deepseek-r1/deepseek-v3 model #1251

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 3 commits into from
Aug 4, 2025
Merged

feat: add deepseek-r1/deepseek-v3 model #1251

merged 3 commits into from
Aug 4, 2025

Conversation

zhuangqh
Copy link
Collaborator

Reason for Change:

Requirements

  • added unit tests and e2e tests (if applicable).

Issue Fixed:

Notes for Reviewers:

@zhuangqh zhuangqh requested review from Fei-Guo and chewong as code owners July 10, 2025 00:05
@kaito-pr-agent Kaito PR Agent
Copy link

Title

Add support for deepseek-r1 and deepseek-v3 models

Description

  • Added support for deepseek-r1 and deepseek-v3 models

  • Registered new models in plugin registration

  • Defined constants and runtime parameters for new models

  • Updated supported_models.yaml with new models

Changes walkthrough 📝

Relevant files
Enhancement
model.go
Add deepseek-r1 and deepseek-v3 support                                   

presets/workspace/models/deepseek/model.go

  • Registered deepseek-r1 and deepseek-v3 models in plugin registration
  • Defined constants for new models
  • Added runtime parameters for deepseek-r1 and deepseek-v3
  • Implemented GetInferenceParameters, GetTuningParameters,
    SupportDistributedInference, and SupportTuning methods for new models
    		• +101/-0 
    Configuration changes
    supported_models.yaml
    Update supported models                                                                   

    presets/workspace/models/supported_models.yaml

    • Added entries for deepseek-r1-0528 and deepseek-v3-0324 models
    		 +13/-0   
    Need help?
  • Type /help how to ... in the comments thread for any questions about PR-Agent usage.
  • Check out the documentation for more information.
    		•

    @kaito-pr-agent Kaito PR Agent
    Copy link

    PR Reviewer Guide 🔍

    Here are some key observations to aid the review process:

    ⏱️ Estimated effort to review: 3 🔵🔵🔵⚪⚪
    🧪 No relevant tests
    🔒 No security concerns identified
    ⚡ Recommended focus areas for review

    Duplicate Metadata

    The Metadata field in the GetInferenceParameters method for both deepseekR1 and deepseekV3 uses PresetDeepSeekR1Model. This might be incorrect and should be PresetDeepSeekV3Model for deepseekV3.

     
    return &model.PresetParam{
    Inconsistent Model Names

    The ModelName field in the VLLM struct for both deepseekR1 and deepseekV3 uses PresetDeepSeekR1Model. This should be consistent with the actual model names PresetDeepSeekR1Model and PresetDeepSeekV3Model.

     
    ModelRunParams:       deepseekR1RunParamsVLLM,

    @kaito-pr-agent Kaito PR Agent
    Copy link

    PR Code Suggestions ✨

    Explore these optional code suggestions:

    CategorySuggestion                                                                                                                                    Impact
    Possible issue
    Correct preset name

    Use the correct preset name for deepseekV3.

    presets/workspace/models/deepseek/model.go [194]

    -Metadata:                  metadata.MustGet(PresetDeepSeekR1Model),
+Metadata:                  metadata.MustGet(PresetDeepSeekV3Model),
    Suggestion importance[1-10]: 8

    __

    Why: The Metadata field for deepseekV3 incorrectly uses PresetDeepSeekR1Model. This should be corrected to PresetDeepSeekV3Model to match the intended model registration.

    		Medium

    @chewong chewong added this to the 0.6.0 milestone Jul 10, 2025
    chewong
    chewong reviewed Jul 10, 2025
    View reviewed changes
    @zhuangqh zhuangqh moved this to In Progress in KAITO Roadmap Jul 15, 2025
    @zhuangqh
    feat: add deepseek-r1/deepseek-v3 model
    3c2687c 
    Signed-off-by: zhuangqh <zhuangqhc@gmail.com>
    @zhuangqh
    fix
    6d83a42 
    Signed-off-by: zhuangqh <zhuangqhc@gmail.com>
    @Fei-Guo Fei-Guo merged commit dc46418 into kaito-project:main Aug 4, 2025
    18 of 20 checks passed
    @github-project-automation github-project-automation bot moved this from In Progress to Done in KAITO Roadmap Aug 4, 2025
    Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
    Reviewers

    @chewong chewong chewong approved these changes

    @Fei-Guo Fei-Guo Fei-Guo approved these changes

    Assignees

    @zhuangqh zhuangqh

    Labels
    Review effort 3/5
    Projects
    KAITO Roadmap
    Status: Done
    Milestone
    0.7.0
    Development

    Successfully merging this pull request may close these issues.
    3 participants
    @zhuangqh @chewong @Fei-Guo