所有应用程序
开发者工具

Immich

使用 Immich、支持矢量的私有数据库和机器学习工作者备份照片和视频。

持久存储

应用程序数据跨部署保留在项目范围的卷上。

已发布的图片

安装固定的上游容器,无需在服务器上构建映像。

Photos and videos on your server

Immich backs up and organizes photos and videos. This template follows the pinned release's upstream stack: matching server and machine-learning images, Valkey, and the supported vector-enabled PostgreSQL image. No customer-side image build is required.

First login

Choose the initial administrator's email and display name. Openship generates and encrypts the password. Registration runs against a loopback-only startup listener; public access starts only after the owner exists, with unauthenticated setup disabled. Changed account details and passwords survive redeployment. Retrieve the initial credentials in Connection and use the same server address in Immich's mobile app.

Requirements and storage

Provide at least 2 CPU cores and 6 GiB RAM; Immich recommends 8 GiB or more. Current AMD64 machine-learning images require an x86-64-v2 CPU. ARM64 is supported. The template uses software processing, without GPU devices or privileged container access. Recommendations do not impose container limits.

Photos, database contents, and downloaded machine-learning models have separate persistent volumes. Keep the database on local Linux-compatible storage, preferably SSD. Model downloads require internet access and may take time when processing the first photos. Back up the database and uploaded media together; this installation is not a substitute for a separate backup.

Official requirements · Release stack · Mobile setup

包含在此模板中

由目录配置的服务作为一个项目一起部署。

immich-db

ghcr.io/immich-app/postgres:14-vectorchord0.4.3-pgvectors0.2.0@sha256:bcf63357191b76a916ae5eb93464d65c07511da41e3bf7a8416db519b40b1c23

immich-broker

valkey/valkey:9@sha256:c123e3715db63d06d4ad6964884037aa0d5d4d703939b9929954112889708e1d

immich-machine-learning

ghcr.io/immich-app/immich-machine-learning:v3.3.1

immich

ghcr.io/immich-app/immich-server:v3.3.1