README 未给出安装步骤,需先查阅 get_started.md 文档;需配合 X-AnyLabeling 客户端使用,模型需自行按插件方式接入

X-AnyLabeling-Server
CVHub520AGPL-3.0活跃维护为 X-AnyLabeling 标注工具提供轻量可扩展的模型推理服务,把检测、分割、分类、姿态等视觉模型接成后端,方便标注时自动预标注。
A Simple, Lightweight, and Extensible Serving Framework for X-AnyLabeling
README 未给出安装命令,详见 docs/source/en/get_started.mdREADME 未说明
为 X-AnyLabeling 提供轻量可扩展的模型推理服务框架,让自动标注模型快速上线
为 X-AnyLabeling 标注工具提供轻量可扩展的模型推理服务,把检测、分割、分类、姿态等视觉模型接成后端,方便标注时自动预标注。
核心能力
- 框架与模型实现解耦,只管服务管理与资源调度
- 插件式架构,不改核心代码即可接入自定义模型
- 内置结构化日志、错误处理、并发控制与安全认证
- 所有参数可配置且带合理默认值,适配不同部署场景
容易踩的坑
- v0.0.15最新正式版2026-11-04 · 2026-11-04 17:13 UTC
🚀 New Features
- Add SAM 3.1 multiplex video segmentation support
- Add RelateAnything open-vocabulary relation recognition support
🐛 Bug Fixes
- Refresh the OpenAPI schema for document prediction
🌟 Contributors
A total of 1 developer contributed to this release.
Thank @CVHub520
- v0.0.142026-10-30 · 2026-10-30 11:55 UTC
查看更新说明
🚀 New Features
- Add Unlimited-OCR-2 document parsing with Transformers and SGLang backends
🌟 Contributors
A total of 1 developer contributed to this release.
Thank @CVHub520
- v0.0.12X-AnyLabeling-Server v0.0.122026-08-05 · 2026-08-05 16:23 UTC
查看更新说明
🐛 Bug Fixes
- Fix remote YOLO class filtering across detection, segmentation, pose, OBB, and tracking
- Fix Segment Anything 2 device handling and concurrent prompt inference (#39)
- Declare the packaging dependency required by the update checker
🛠️ Improvements
- Organize bilingual documentation and automate API reference generation
- Validate OpenAPI dependencies in CI
- Update the recommended project citation
🌟 Contributors
A total of 1 developer contributed to this release.
Thank @CVHub520
Full Changelog: https://github.com/CVHub520/X-AnyLabeling-Server/compare/v0.0.11...v0.0.12
- v0.0.11X-AnyLabeling-Server v0.0.112026-06-06 · 2026-06-06 13:58 UTC
查看更新说明
🎬 Visual Demo
Dense Object Detection LocateAnything performs diverse localization tasks under a unified VLM — document understanding, GUI grounding, dense object detection, and OCR.
Fast Decoding Speed Parallel Box Decoding (PBD) vs. Quantized Coordinate Decoding — PBD predicts each bounding box atomically in a single forward pass for substantially faster throughput.
Link: https://github.com/CVHub520/X-AnyLabeling/blob/main/examples/grounding/locateanything/README.md
🚀 New Features
- Add LocateAnything support for remote grounding, pointing, and text detection
🌟 Contributors
A total of 1 developer contributed to this release.
Full Changelog: https://github.com/CVHub520/X-AnyLabeling-Server/compare/v0.0.10...v0.0.11
- v0.0.10X-AnyLabeling-Server v0.0.102026-04-25 · 2026-04-25 07:21 UTC
查看更新说明
v0.0.10(Apr 25, 2026)🚀 New Features
- Add reviewed YOLOE 26s segmentation configs
🐛 Bug Fixes
- Use configured CUDA device string for SAM3 inference
🌟 Contributors
A total of 3 developers contributed to this release.
Thank @93054182, @rycerzes, @CVHub520
用户点评
登录用户的真实使用记录,可按版本和硬件筛选还没有人点评。第一个说出真实体验的人最有价值。

About
X-AnyLabeling-Server is a simple, lightweight and extensible serving framework for AI model inference, specifically designed for X-AnyLabeling. It provides a production-ready solution with pluggable architecture and flexible configuration for various auto-labeling scenarios. Its core features include:
- Decoupled Design: Framework handles service management and resource scheduling without interfering with model implementation details
- Pluggable Architecture: Rapidly integrate custom models without modifying core framework code
- Production Ready: Comprehensive structured logging, error handling, concurrency control, and security authentication
- Flexible Configuration: All parameters are configurable with sensible defaults, adaptable to different deployment scenarios
Getting Started
Contributing
Contributions and collaborations are highly appreciated! For guidelines on how to contribute, please refer to Contributing to X-AnyLabeling-Server.
Sponsors
Citing
If you use this software in your research, please cite it as below:
@misc{X-AnyLabeling-Server,
year = {2025},
author = {Wei Wang},
publisher = {Github},
organization = {CVHub},
journal = {Github repository},
title = {A Simple, Lightweight, and Extensible Serving Framework for X-AnyLabeling},
howpublished = {\url{https://github.com/CVHub520/X-AnyLabeling-Server}}
}
@misc{X-AnyLabeling,
year = {2023},
author = {Wei Wang},
publisher = {Github},
organization = {CVHub},
journal = {Github repository},
title = {X-AnyLabeling: A Unified Desktop Platform for AI-Assisted Data Annotation
},
howpublished = {\url{https://github.com/CVHub520/X-AnyLabeling}}
}