xTag

Overview

Full name
docker.io/openmmlab/lmdeploy
Registry
docker.io
Namespace
openmmlab
Repository
lmdeploy
Cached tags
140
Last synced
07/13/2026, 12:39 PM

Introduction

LMDeploy 大模型推理部署镜像，TurboMind 引擎高吞吐 LLM serving，OpenMMLab 系列模型 GPU 推理与服务化部署工具。

Details

openmmlab/lmdeploy 集成 LMDeploy 推理框架与 TurboMind 引擎，支持 Llama、InternLM 等模型的高效 GPU 推理与 OpenAI 兼容 API。定位为 GPU LLM 推理服务，需 NVIDIA GPU；私有化大模型 API 网关、InternLM 部署或替代 vLLM 的高性能推理方案。

快速启动

docker run -d --name lmdeploy \
  --gpus all \
  -p 23333:23333 \
  docker.io/openmmlab/lmdeploy:latest \
  lmdeploy serve api_server internlm2-chat-7b

推荐实践

docker run -d --name lmdeploy \
  --gpus all \
  --restart unless-stopped \
  -p 23333:23333 \
  -v /data/models:/models \
  docker.io/openmmlab/lmdeploy:v0.6.0 \
  lmdeploy serve api_server /models/internlm2 --tp 2

核心参数说明

  • --gpus all — NVIDIA GPU 必需
  • -p 23333 — OpenAI 兼容 API 端口
  • --tp 2 — 张量并行 GPU 数
  • -v .../models:... — 本地模型权重

Latest tags (20)

View all 140 tags →
TagArchPushed AtSizeDigestSyncWishAction
docker.io/openmmlab/lmdeploy
:nightly-cu12.8
linux/amd64
2026/07/13 20:165.2 GBsha256:bc9d51fb9a2b…
docker.io/openmmlab/lmdeploy
:nightly-test-cu12.8
linux/amd64
2026/07/13 19:505.8 GBsha256:65904644d8e8…
docker.io/openmmlab/lmdeploy
:v0.14.0-cu12.8
linux/amd64
2026/06/24 14:125.2 GBsha256:e5f59ec579ec…
docker.io/openmmlab/lmdeploy
:latest-cu12.8
linux/amd64
2026/06/24 14:125.2 GBsha256:e5f59ec579ec…
docker.io/openmmlab/lmdeploy
:v0.14.0-cu12
linux/amd64
2026/06/24 13:575.2 GBsha256:e7091fdaedcf…
docker.io/openmmlab/lmdeploy
:latest
linux/amd64
2026/06/24 13:575.2 GBsha256:e7091fdaedcf…
docker.io/openmmlab/lmdeploy
:latest-cu12
linux/amd64
2026/06/24 13:575.2 GBsha256:e7091fdaedcf…
docker.io/openmmlab/lmdeploy
:ascend
linux/arm64
2026/06/24 13:424.4 GBsha256:1a4f689aab3b…
docker.io/openmmlab/lmdeploy
:v0.14.0a2
linux/amd64
2026/06/16 12:295.2 GBsha256:32b51a4b1e0d…
docker.io/openmmlab/lmdeploy
:v0.14.0a1
linux/amd64
2026/06/01 17:205.2 GBsha256:9cb5ad8466e3…
docker.io/openmmlab/lmdeploy
:nightly-test-cu12.8_v0.14.0a0
linux/amd64
2026/05/21 21:385.8 GBsha256:32ca87cdeabb…
docker.io/openmmlab/lmdeploy
:nightly-cu12.8_v0.14.0a0
linux/amd64
2026/05/21 21:365.2 GBsha256:33d77c6f43c3…
docker.io/openmmlab/lmdeploy
:v0.13.0-cu12.8
linux/amd64
2026/05/12 13:045.1 GBsha256:df9caee47a68…
docker.io/openmmlab/lmdeploy
:v0.13.0-cu12
linux/amd64
2026/05/12 12:335.1 GBsha256:37a5e38df2a6…
docker.io/openmmlab/lmdeploy
:nightly-test-cu12.8_v0.13.0_2
linux/amd64
2026/05/07 18:475.8 GBsha256:ccc2542e6dd1…
docker.io/openmmlab/lmdeploy
:nightly-cu12.8_v0.13.0_2
linux/amd64
2026/05/07 18:425.1 GBsha256:8e80c5ac067c…
docker.io/openmmlab/lmdeploy
:nightly-test-cu12.8_v0.13.0
linux/amd64
2026/04/29 20:585.8 GBsha256:530b1cdc3a15…
docker.io/openmmlab/lmdeploy
:nightly-cu12.8_v0.13.0
linux/amd64
2026/04/29 20:565.1 GBsha256:780f5665fa20…
docker.io/openmmlab/lmdeploy
:v0.12.3-a3
linux/arm64
2026/04/20 15:274.4 GBsha256:d232ccc99ca9…
docker.io/openmmlab/lmdeploy
:v0.12.3.post2-cu12.8
linux/amd64
2026/04/09 18:435.2 GBsha256:2558bbdd1240…