Everybody states Immich is the very best Google Photos option, however this complimentary self-hosted option has much more functions

Science

science 4

Released Oct 8, 2026, 1:30 PM EDT

Gregory Gibson is an IT expert and technical author who concentrates on Linux and Windows Servers, Android modification, virtual personal servers, web servers, and cybersecurity.

Gregory has actually operated in business IT for over 15 years and delights in developing content to assist individuals get the most out of their homelab and self-hosted IT tasks, and to remain safe online.

When not working, he takes pleasure in photography, composing sci-fi, and digital art.

The challenging part about discovering a Google Photos replacement isn’t keeping pictures or working within its community. Typically it’s simply discovering an option that makes pictures simply as simple to browse, arrange, and share as Google does.

Immich currently covers much of that, that made me question whether Noodle Gallery might enhance on it.

Noodle Gallery is a self-hosted Immich fork with loads of additional functions for handling and sharing an image library. I installed it on among my Docker VMs and published simply over 120 images to see if the functions made it an excellent option to Google Photos.

The familiarity of Immich made an excellent impression, however the additionals have me well and genuinely connected.

The picture library made a strong impression

Browse discovered particular vehicles, and my collection was simple to search on both screens

I set up Noodle Gallery v5.7.1 through Docker Compose on a Proxmox virtual device with 2 CPU cores, 8 GB of RAM, and a 50 GB disk. That was my test setup, and it had adequate space and resources to handle background processing jobs.

As soon as the pictures were published, I might search them by date, open private images, and check the image EXIF information. I likewise collected all the seaside photos and popped them into an album called “Ocean.” I then set up the Noodle Gallery Android app and linked it to my server, and might see the collection from either screen.

All of this is typical habits I anticipated from an image company app, however Search was where I began having a good time.

Among the very best functions of Google Photos is its image search. You can look for the most granular terms, and in some way it produces a collection of your own images that match them with frightening precision. Immich did a great task of reproducing this, however its constraints ended up being apparent as soon as you began utilizing more particular search terms.

Naturally, this is among the very first things I evaluated with Noodle Gallery. Searching for a picture of a “feline” would have been a simple very first test, however I wished to know how particular I might get without manual labeling:

  • “Seal point feline” raised photos of my feline and left out felines with other pigmentation patterns.

  • “Lada Niva” and “Suzuki Jimny” raised pictures of those exact designs of cars and trucks.

  • “Snowy mountains” got the winter season scenes particularly with mountains just.

  • “Seagull” selected all the seagulls and left out images of parrots and pigeons.

I had not tagged a single image before I began. Having the ability to type what I kept in mind about a photo made the Noodle Gallery quite close in regards to Google Photos’ search performance.

Noodle does acquire this ability and control from Immich, so credit goes where credit is due. The UI itself was much snappier and faster than my previous experience with Immich running on the very same resources and dealing with comparable images.

Noodle Gallery

RAM

4GB minimum. 6GB or more for artificial intelligence

Platform(s)

Web, Android, iOS

Designer

Open Noodle and neighborhood factors, developed on Immich

Cost design

Free and open-source

It acknowledged my feline before he matured

The kittycat image made family pet acknowledgment feel really helpful

It matched my feline Jarvis ‘kitty picture and kept the other felines out. That a person function captured my attention more than any other in the app’s description. Jarvis the feline has actually because died, so seeing these photos combined various phases of his life and offered the collection of images that Noodle emerged some individual weight.

I ‘d allowed its family pet detection function, which works like facial acknowledgment, just for animals. I let the processing surface, looked under Explore, and called the

resulting animal entry “Jarvis.”science Sunsets Shared Space in Noodle Gallery displaying a shared collection of sunset photos Opening it provided me a collection of his images without ever requiring to pick every one separately and by hand include it to an album.

Searching for “feline” would’ve discovered all the felines in my pictures(and there are a great deal of them! ), however Noodle’s family pet detection function acknowledged this one specific feline throughout noticeably various ages.

It matched my feline Jarvis’kitty picture with his adult images and kept the other felines out.

Now, Jarvis was a really unique-looking feline, so I can’t guarantee Noodle’s family pet detection will differentiate every family pet similarly well. Still, the properly organized pictures revealed why I and other animal owners would desire the function. It assisted me review images with individual significance without needing to by hand sort through an unpleasant image library.

The other account’s sundowns appeared right away along with mine

Integrating pictures from various accounts is uncomfortable when I just wish to share part of each collection with other relative or good friends. I wished to see whether Noodle Gallery’s Shared Spaces might bring photos together while still leaving the rest of each library different.

This function comes in handy for when

you have several individuals taking pictures at an occasion, state, a wedding event, and you wish to integrate everybody’s images into a single album.

I developed a Space called “Sunset Photography,” included pictures from my primary account, and after that provided a 2nd test account science Noodle Gallery screenshot classification results containing screenshots and incorrectly tagged photographs of screens Editor gain access to. From that test account, I submitted another choice of sunset pictures and included them to the Space.

From the primary account’s point of view, the images appeared instantly. Much better still, they slotted nicely into my existing pictures purchased by date taken. I didn’t require to download the other account’s images, nor upload another copy. The shared collection was currently there, set up into a single unified timeline.

Consents considered that shared area some structure:

  • The Owner handles the area and its subscription.

  • An Editor can contribute and handle pictures in the collection.

  • A Audience can search without the capability to alter the images.

I might likewise inspect the activity log to see which account had actually included which pictures. That made it far simpler to track modifications, particularly compared to doing it by memory, which left space for a lot of errors.

Activity log for the Sunsets Shared Space showing photo contributions from two test accounts

Immich currently supports shared albums, so once again, sharing isn’t a novelty that’s special to Noodle. What did interest me, however, was the mix of a shared timeline, member functions, and trackable activity.

Even with simply 2 test accounts, I might see myself encouraging loved ones to join my Noodle server so we might gather everybody’s images from shared journeys and events in one location.

Automatic arranging required a little guidance

It discovered my screenshots, however pictures of screens deceived it too

Noodle’s search performance was an immediate hit for me, however automated category still required a little bit of the human touch to work effectively. Gallery lets me explain a classification in plain language, then utilizes OpenAI’s open-source CLIP (Contrastive Language-Image Pre-training) to tag pictures that fit. This function is distinct to Noodle Gallery, and among the primary factors I wished to attempt it. It can likewise archive matches instantly, though I selected Tag just I might at least check its choices.

My preliminary descriptions covered screenshots of phone screens, computer system applications, and sites. By default, the resemblance limit is set to 0.28, and utilizing this, the Gallery discovered just 2 Windows-related images science Noodle Gallery Spaces overview showing the populated Sunsets shared collection and missed out on the screenshots I submitted particularly for the test.

Reducing that limit to 0.24 generated all the desired screenshots. The outcomes likewise consisted of pictures of displays and my automobile’s stereo LCD screen.

Setup evaluated

Lead to my collection

Preliminary triggers, 0.28 resemblance

2 matches, missing out on the test screenshots

Broadened triggers, 0.24 resemblance

All test screenshots, plus pictures of screens

Broadened triggers, 0.25 resemblance

The majority of test screenshots were missed out on

Increasing the limit by 0.01 to 0.25 made the matching too rigorous for my collection. I stopped there and kept the more comprehensive outcomes for a manual evaluation. The classifier acknowledged screen-like material, however sadly, it could not separate in between an image of a screen and a screenshot.

Screenshot classification settings in Noodle Gallery showing custom prompts and a similarity threshold of 0.24

Tagging still assisted collect possible prospects, however automated archiving is simply too difficult to validate making it possible for when common pictures might get swept up too.

Enable Tags under Account Settings → Functions to search produced tags. Start with Tag justthen examine the outcomes. If you’re delighted with its choice, switch on automated archiving.

More functions are just great when they fix issues I can really see

Noodle Gallery Spaces overview showing the populated Sunsets shared collection

Immich has a much bigger neighborhood and a longer performance history. Those are 2 excellent factors to stay with it, specifically if you currently have a recognized image library. Noodle Gallery develops on the great Immich has actually done, however including functions does not immediately make the fork worth changing to.

My test likewise had its limitations. I just utilized a little collection of test pictures, and I still required to examine the category results. I likewise had not moved my whole image library of countless images throughout or checked whether I ‘d totally rely on Noodle Gallery with years of backups.

In stating that, my old image management option, Google Photos, isn’t a backup tool either. I would easily move my own images over to either Immich or Noodle, offered a secondary image backup strategy remained in location.

Still, the additions offered me truly beneficial outcomes. My feline’s kitty and adult images all came together, the shared sundowns appeared instantly, and the classifier collected my screenshots when I ‘d modified the settings. Every brand-new function actually did something I might value while searching my own images.

Truthfully, that’s sufficient for me to keep utilizing Noodle Gallery and see how it holds up when moving a genuinely huge image collection. If add-on functions like animal detection and prompt-based automated category do not actually have an usage for you, then Immich might currently be whatever you require.

In my viewpoint, Open-Noodle has actually made the fork worth investing more time on, and in the meantime, produced a definitely practical option to Google Photos.

Learn more


Discover more from PMN S.P.O.R.T.S - A PRIME MEDIA NETWORK BRAND

Subscribe to get the latest posts sent to your email.

Related Articles

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Captcha verification failed!
CAPTCHA user score failed. Please contact us!