Skip to main content

A lightweight, effective and easy-to-extend inference runtime

Project description

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

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]

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

xlite-0.1.0rc10-cp312-cp312-manylinux2014_x86_64.whl (794.5 kB view details)

Uploaded CPython 3.12

xlite-0.1.0rc10-cp312-cp312-manylinux2014_aarch64.whl (749.7 kB view details)

Uploaded CPython 3.12

xlite-0.1.0rc10-cp311-cp311-manylinux2014_x86_64.whl (771.3 kB view details)

Uploaded CPython 3.11

xlite-0.1.0rc10-cp311-cp311-manylinux2014_aarch64.whl (736.4 kB view details)

Uploaded CPython 3.11

File details

Details for the file xlite-0.1.0rc10-cp312-cp312-manylinux2014_x86_64.whl.

File metadata

File hashes

Hashes for xlite-0.1.0rc10-cp312-cp312-manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 3e4d58db6989142766eb311c71d53164d2d2e77f96302b2f414318f07d873629
MD5 3e9d00a6b05f87d338d172435f3dd4a5
BLAKE2b-256 cd55b2443cd223ee2f2b9d58f35818bc1225e261ebbde024e892d855a1986825

See more details on using hashes here.

File details

Details for the file xlite-0.1.0rc10-cp312-cp312-manylinux2014_aarch64.whl.

File metadata

File hashes

Hashes for xlite-0.1.0rc10-cp312-cp312-manylinux2014_aarch64.whl
Algorithm Hash digest
SHA256 9eef6c71a0d70cf6b7de786da21257ca6455533479209bce8ca8c57479af75b1
MD5 2f8b077c1cf0e3a7c7203d80fb75c46f
BLAKE2b-256 f7b6a508925fe6ddad9b320e30b9e25f5e104fb551ae45b7e84ff57ae89d68af

See more details on using hashes here.

File details

Details for the file xlite-0.1.0rc10-cp311-cp311-manylinux2014_x86_64.whl.

File metadata

File hashes

Hashes for xlite-0.1.0rc10-cp311-cp311-manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 8ffa4551b9d7d27cbe3e87d8c5f11f8376f0ffee3ede937b0777789b69f94289
MD5 0987c802a8b95a076d1e8f1793799bc8
BLAKE2b-256 4bf9b4af9917d0c5ab49161b42c3987fce85e81cd45ccd388d165eaeab68b9fd

See more details on using hashes here.

File details

Details for the file xlite-0.1.0rc10-cp311-cp311-manylinux2014_aarch64.whl.

File metadata

File hashes

Hashes for xlite-0.1.0rc10-cp311-cp311-manylinux2014_aarch64.whl
Algorithm Hash digest
SHA256 6faab2e63d370ca40d10b8890165777cc7db254988b3791b474a02a1d80b4c73
MD5 e99e584bc8294e2402811266f4bbf69c
BLAKE2b-256 4f676c56893a6b04611e52ef217a40f23b00120a7da493f7414e4e1584f35f9b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page