immich-app/immich
1
0
Fork 0
mirror of https://github.com/immich-app/immich.git synced 2026-07-25 14:00:45 +03:00
Code Issues 783 Packages Projects Releases 22 Wiki Activity

[PR #3797] [MERGED] chore(ml): make lazy loading default #10119

New Issue
Closed
opened 2026-02-05 14:16:19 +03:00 by OVERLORD · 0 comments
OVERLORD commented 2026-02-05 14:16:19 +03:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/immich-app/immich/pull/3797
Author: @mertalev
Created: 8/20/2023
Status: ✅ Merged
Merged: 8/21/2023
Merged by: @alextran1502

Base: main ← Head: chore/ml-lazy-loading

📝 Commits (1)

  • 25bbb71 make lazy loading default

📊 Changes

1 file changed (+1 additions, -1 deletions)

View changed files

📝 machine-learning/app/config.py (+1 -1)

📄 Description

Description

This change means a model will only load when the ML service receives a relevant request. This doesn't change unloading behavior, so models will stay loaded by default. Additionally, models will still be downloaded on startup. The benefit of this change is to optimize RAM usage. Notably, disabling particular ML jobs with #3768 means the corresponding model won't be loaded.

How Has This Been Tested?

I ran docker stats --no-stream after startup, confirming low RAM usage for immich_machine_learning at under 300mb. I then started and stopped one ML job at a time and checked docker stats --no-stream again for each, confirming the expected increase in RAM with each model.

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/immich-app/immich/pull/3797 **Author:** [@mertalev](https://github.com/mertalev) **Created:** 8/20/2023 **Status:** ✅ Merged **Merged:** 8/21/2023 **Merged by:** [@alextran1502](https://github.com/alextran1502) **Base:** `main` ← **Head:** `chore/ml-lazy-loading` --- ### 📝 Commits (1) - [`25bbb71`](https://github.com/immich-app/immich/commit/25bbb710a76c88e08598bb2e5a699ae6c0fd6f8f) make lazy loading default ### 📊 Changes **1 file changed** (+1 additions, -1 deletions) <details> <summary>View changed files</summary> 📝 `machine-learning/app/config.py` (+1 -1) </details> ### 📄 Description ## Description This change means a model will only load when the ML service receives a relevant request. This doesn't change unloading behavior, so models will stay loaded by default. Additionally, models will still be downloaded on startup. The benefit of this change is to optimize RAM usage. Notably, disabling particular ML jobs with #3768 means the corresponding model won't be loaded. ## How Has This Been Tested? I ran `docker stats --no-stream` after startup, confirming low RAM usage for `immich_machine_learning` at under 300mb. I then started and stopped one ML job at a time and checked `docker stats --no-stream` again for each, confirming the expected increase in RAM with each model. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
OVERLORD added the pull-request label 2026-02-05 14:16:19 +03:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
accessibility
changelog:enhancement
changelog:security
changelog:skip
changelog:translation
cli
date-time
dependencies
documentation
external-library
format
good first issue
mobile-beta
mobile-beta
mobile-beta
needs-answer
nice to have
pull-request
Mirrored from GitHub Pull Request
 sharing
tech-debt
📱mobile
🖥️web
🗄️server
🧠machine-learning
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
OVERLORD
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: immich-app/immich#10119
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?