nextcloud / nextcloud/recognize

Recognize maintenance routine unnecessarily redownloads Tensorflow CPU and GPU models every time

Open
#1,478 3 comments 0 reactions 1 assignee View on GitHub

@marcelklehr is already working on this.

Since Apr 1, 2026.

bug
Dominant language
PHP
Stars
699
Forks
68
Avg merge
1d 4h
Merged PRs (30d)
5

Description

Which version of recognize are you using?

11.0.1

Enabled Modes

Face recognition

TensorFlow mode

Normal mode

Downstream App

Memories App

Which Nextcloud version do you have installed?

33.0.1

Which Operating system do you have installed?

Debian 13

Which database are you running Nextcloud on?

MariaDB 11.8.6

Which Docker container are you using to run Nextcloud? (if applicable)

N/A

How much RAM does your server have?

4GiB

What processor Architecture does your CPU have?

x86_64

Describe the Bug

While running occ maintenance:repair I noticed that the Recognize step took a bit of time the command is run. After a bit of debugging, it appears that Recognize is explicitly forcing a re-download of both the CPU (tfjs-node) and GPU (tfjs-node-gpu) models every time which seems completely unnecessary. It also does this regardless of whether occ maintenance:repair included the --include-expensive option.

Expected Behavior

The Recognize maintenance step should only perform necessary maintenance and not re-download ~150 MB of dependencies.

To Reproduce

Run occ maintenance:repair

Debug log

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.