Skip to content

[RadixTree][3/N Refactor]：Support unified insert/evict params#17401

Merged
ispobock merged 2 commits intosgl-project:mainfrom
hzh0425:hybrid/unified_insert_evict
Jan 22, 2026
Merged

[RadixTree][3/N Refactor]：Support unified insert/evict params#17401
ispobock merged 2 commits intosgl-project:mainfrom
hzh0425:hybrid/unified_insert_evict

Conversation

@hzh0425
Copy link
Copy Markdown
Collaborator

@hzh0425 hzh0425 commented Jan 20, 2026 •
edited
Loading

Motivation

Modifications

Accuracy Tests

qwen-next mamba

Accuracy: 0.952
Invalid: 0.000
Latency: 16.043 s
Output throughput: 5001.579 token/s

qwen 32b

Accuracy: 0.940
Invalid: 0.000
Latency: 25.449 s
Output throughput: 2838.522 token/s

llama4 swa

Accuracy: 0.924
Invalid: 0.000
Latency: 49.150 s
Output throughput: 1037.788 token/s

Benchmarking and Profiling

Checklist

Review Process

  1. Ping Merge Oncalls to start the PR flow. See the PR Merge Process.
  2. Get approvals from CODEOWNERS and other reviewers.
  3. Trigger CI tests with comments or contact authorized users to do so.
    • /tag-run-ci-label, /rerun-failed-ci, /tag-and-rerun-ci
  4. After green CI and required approvals, ask Merge Oncalls to merge.

@hzh0425 hzh0425 added the run-ci label Jan 20, 2026
@gemini-code-assist
Copy link
Copy Markdown
Contributor

gemini-code-assist Bot commented Jan 20, 2026

Summary of Changes

Hello @hzh0425, 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 refactors the RadixTree-based memory cache system by introducing unified parameter and result objects for insert and evict operations. This change aims to standardize the API for cache management across various specialized cache implementations (e.g., Mamba, SWA, Chunked), improving code consistency, readability, and maintainability. By encapsulating operation-specific details within dedicated dataclasses, the refactoring simplifies method signatures and prepares the system for future extensions or modifications to cache behavior.

Highlights

  • Unified Parameter Objects: Introduced InsertParams and EvictParams dataclasses to standardize the arguments passed to insert and evict methods across different cache implementations.
  • Standardized Return Objects: Introduced InsertResult and EvictResult dataclasses to provide a consistent return type for insert and evict operations, encapsulating relevant outcomes.
  • API Refactoring: Modified the insert and evict methods in BasePrefixCache and its concrete implementations (RadixCache, MambaRadixCache, SwaRadixCache, ChunkCache, HiRadixCache, LmcRadixCache, RadixCacheCpp) to use these new parameter and result objects.
  • Improved Mamba Cache Eviction: The MambaRadixCache's evict_mamba method was integrated into a unified evict method that can handle both full KV cache and Mamba state eviction via EvictParams.
  • Test Case Updates: Corresponding unit tests were updated to reflect the new API signatures and parameter passing for insert and evict methods, including assertions on the new result objects.

🧠 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 Jan 20, 2026
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

The pull request successfully refactors the insert and evict methods across various cache implementations to use unified InsertParams, InsertResult, EvictParams, and EvictResult dataclasses. This improves consistency and maintainability of the cache interfaces. All call sites and method signatures have been updated correctly to reflect these changes, including the test files. The refactoring is well-executed and enhances the clarity of the cache operations.

@hzh0425 hzh0425 force-pushed the hybrid/unified_insert_evict branch from 9d5f322 to 81e2108 Compare January 21, 2026 12:14
@ispobock ispobock merged commit f33022d into sgl-project:main Jan 22, 2026
56 of 94 checks passed
Johnsonms pushed a commit to Johnsonms/sglang that referenced this pull request Feb 14, 2026
@hzh0425 @Johnsonms
[RadixTree][3/N Refactor]：Support unified insert/evict params (sgl-pr…
8922a91 
…oject#17401)
@ispobock ispobock mentioned this pull request Mar 12, 2026
Open
23 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@ispobock ispobock ispobock approved these changes

@hanming-lu hanming-lu hanming-lu approved these changes

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

@Ying1123 Ying1123 Awaiting requested review from Ying1123 Ying1123 is a code owner

@hnyls2002 hnyls2002 Awaiting requested review from hnyls2002 hnyls2002 is a code owner

@xiezhq-hermann xiezhq-hermann Awaiting requested review from xiezhq-hermann xiezhq-hermann is a code owner

@yizhang2077 yizhang2077 Awaiting requested review from yizhang2077 yizhang2077 is a code owner

+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

@yizhang2077 yizhang2077

@ispobock ispobock

@hanming-lu hanming-lu

Labels

run-ci

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@hzh0425 @ispobock @hanming-lu @yizhang2077