Skip to main content
Pre-release

This release is a pre-release and may not be stable for production use.

xlite logo

Xlite轻量化推理运行时

介绍

xlite (GVirt前端):轻量级Transformer模型运行时,支持多样性算力协同,当前支持在昇腾硬件上高效运行。 xlite公开了Transformer运行所需的模型构图以及算子,所有算子基于昇腾AscendC/CCE开发。

背景与动机

在大模型推理场景中,传统的单流串行执行模式存在以下问题:

  • 核间负载不均:不同AICORE之间的任务分配不均衡,部分核心闲置
  • 资源浪费多:计算资源和传输资源利用率低,存在显著浪费
  • 执行时间长:Host CPU下发算子开销大,造成严重的host bond问题

xlite通过以下技术手段解决上述问题:

  • 多流并行:充分利用卡内资源,将单流串行改为多流并行执行
  • 核间负载均衡:核间负载均衡,提升资源利用率
  • CPU NPU协同:C++侧完全消除Python的GC、线程等干扰,简化Host tiling计算,去除小块内存申请释放及拷贝,消除Host bond

性能效果显著:

在GLM-4.7双机推理场景(40K输入、1K输出、prefix cache命中率约90%)下:

  • TPOT时延降低17%~30%
  • 吞吐提升13%~41%

详细性能数据参考 PR #7935。

软件架构

image

xlite已适配vllm_ascend,可通过xlite_graph_config配置快速使能xlite加速效果,使用方法参考官方指导文档。

xlite支持模型见模型列表。

快速开始

安装

# 安装vllm_ascend, 可参考https://github.com/vllm-project/vllm-ascend/blob/main/README.md
# 安装xlite
pip install xlite --extra-index https://download.pytorch.org/whl/cpu/

离线推理示例

import os
from vllm import LLM

# xlite默认支持decode-only模式, 可通过设置 "full_mode": True 使能full模式
model = LLM(model="path/to/Qwen3-32B", tensor_parallel_size=8, additional_config={"xlite_graph_config": {"enabled": True, "full_mode": True}})
outputs = model.generate("Hello, how are you?")

在线服务示例

vllm serve path/to/Qwen3-32B --tensor-parallel-size 8 --additional-config='{"xlite_graph_config": {"enabled": true, "full_mode": true}}'

开发指南

编译构建、容器镜像、源码安装等开发相关内容,请参考 开发指南

环境变量

xlite使用的环境变量及其配置方法,请参考 环境变量

目录结构

.
├── csrc/    - 轻量化运行时的核心代码
├── xlite/   - python代码,包括tools等
├── doc/     - 相关文档介绍
├── docker/  - 容器镜像Dockerfile
└── tests/   - 测试用例

致谢

本项目的核心算子初始版本和思路来自于华为终端小艺AI Infra团队的贡献,相关优化实现可参考论文:

《XY-Serve: End-to-End Versatile Production Serving for Dynamic LLM Workloads》 [ASPLOS 2026]

Metadata

Release files for xlite 0.2.0rc3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for xlite 0.2.0rc3
File Size Uploaded Provenance Actions
xlite-0.2.0rc3.tar.gz 381.4 kB Without provenance Details for xlite-0.2.0rc3.tar.gz

Built distributions (wheels)

Table of built distributions (wheels) for xlite 0.2.0rc3
File Actions
xlite-0.2.0rc3-cp312-cp312-manylinux2014_x86_64.whl CPython 3.12 CPython 3.12 Linux glibc 2.17+ x86-64 Details for xlite-0.2.0rc3-cp312-cp312-manylinux2014_x86_64.whl
xlite-0.2.0rc3-cp312-cp312-manylinux2014_aarch64.whl CPython 3.12 CPython 3.12 Linux glibc 2.17+ ARM64 Details for xlite-0.2.0rc3-cp312-cp312-manylinux2014_aarch64.whl
xlite-0.2.0rc3-cp311-cp311-manylinux2014_x86_64.whl CPython 3.11 CPython 3.11 Linux glibc 2.17+ x86-64 Details for xlite-0.2.0rc3-cp311-cp311-manylinux2014_x86_64.whl
xlite-0.2.0rc3-cp311-cp311-manylinux2014_aarch64.whl CPython 3.11 CPython 3.11 Linux glibc 2.17+ ARM64 Details for xlite-0.2.0rc3-cp311-cp311-manylinux2014_aarch64.whl

Total release size: 5.4 MB

Release files / xlite-0.2.0rc3.tar.gz

Download URL xlite-0.2.0rc3.tar.gz
Size 381.4 kB
Tags Source
SHA-256 checksum
How to use checksums
08905d98f3b56766abdd4a2edaf9c70bdaf651576c4579c005ed00cc86b1780e
BLAKE2b-256 checksum
How to use checksums
9eca00560521da80ebe6b83c73abef9477be6b3bb5ce1d26cb088b7622d8d22b
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Release files / xlite-0.2.0rc3-cp312-cp312-manylinux2014_x86_64.whl

Download URL xlite-0.2.0rc3-cp312-cp312-manylinux2014_x86_64.whl
Size 1.3 MB
Tags CPython 3.12 Linux glibc 2.17+ x86-64
SHA-256 checksum
How to use checksums
a45eecd8b86eef26b8f3fb0655cf48106ab22548b2398546fd756b3b250ea89a
BLAKE2b-256 checksum
How to use checksums
3eaafca9aaf34b2cd05bdbc0bf377ecb73bd3c8fd7f1b75828fdb77bf745fe88
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Release files / xlite-0.2.0rc3-cp312-cp312-manylinux2014_aarch64.whl

Download URL xlite-0.2.0rc3-cp312-cp312-manylinux2014_aarch64.whl
Size 1.2 MB
Tags CPython 3.12 Linux glibc 2.17+ ARM64
SHA-256 checksum
How to use checksums
ff943ec1888045f8446919a06f5b8064ce527913147752658920edde0de373bb
BLAKE2b-256 checksum
How to use checksums
b156b015fd7a6cac4148488e6d137557c4b3d2cf6d7d4d6b3ba7947c7559efe0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.11

Release files / xlite-0.2.0rc3-cp311-cp311-manylinux2014_x86_64.whl

Download URL xlite-0.2.0rc3-cp311-cp311-manylinux2014_x86_64.whl
Size 1.3 MB
Tags CPython 3.11 Linux glibc 2.17+ x86-64
SHA-256 checksum
How to use checksums
3664caaff0540e8ad39155859a26d2b2d566222acf3bd9a76f910d7d4e834eb2
BLAKE2b-256 checksum
How to use checksums
13e35b7b13830eec5b78bc0bf2656348f82764cdba81178c4d352d440f013cc0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Release files / xlite-0.2.0rc3-cp311-cp311-manylinux2014_aarch64.whl

Download URL xlite-0.2.0rc3-cp311-cp311-manylinux2014_aarch64.whl
Size 1.2 MB
Tags CPython 3.11 Linux glibc 2.17+ ARM64
SHA-256 checksum
How to use checksums
56936f722ded334c0b7f120ccada07c6abdfb444ebdab8fe50bd7a8e9cd1a1c4
BLAKE2b-256 checksum
How to use checksums
d28f02e47bfab360a9601282bd66dba629caf2e8f725e42c5998bdbe86904e35
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.11

Security for xlite 0.2.0rc3

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release
Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page