OK, indexing finished some time yesterday and I ran a few searches like:
“Child wearing glasses indoors”
“Cars with no wheels”
“Woman riding a bike”
Results come up (immich on android) in three seconds.
But the quality of the results do appear to be considerably better with ViT-B-16-SigLIP2__webli compared to the default model.
I’m pretty happy. 👍
Search speed was never an issue before, and neither was quality. My biggest gripe is not being able to sort search by date! If I had that, it would be perfect.
But I’ll update you once it’s done (at 97,000 to go… )
I decided on the ViT-B-16-SigLIP2__webli model, so switched to that last night. I also needed to update my server to the latest version of Immich, so a new smart search job was run late last night.
Out of 140,000+ photos/videos, it’s down to 104,000 and I have it set to 6 concurrent tasks.
I don’t mind it processing for 24h. I believe when I first set immich up, the smart search took many days. I’m still able to use the app and website to navigate and search without any delays.
Seemed to be the most popular. LOL The smart search job hasn’t been running for long, so I’ll check that other one out and see how it compares. If it looks better, I can easily use that.
Running now.
That’s a Celeron right?
Yup, the Intel J4125 Celeron 4-Core CPU, 2.0-2.7Ghz.
I switched to the ViT-SO400M-16-SigLIP2-384__webli model, same as what you use. I don’t worry about processing time, but it looks like a more capable model, and I really only use immich for contextual search anyway, so that might be a nice upgrade.
Is this something that would be recommended if self-hosting off a Synology 920+ NAS?
My NAS does have extra ram to spare because I upgraded it, and has NVME cache 🤗
No one follows the rules
This is why ordinary people should feel a duty to use adblockers, poison AI/data collection whenever possible, and to use alternative methods of protecting privacy (i.e. pirate rather than using streaming services).
If we can’t be respected, then we should give no respect to organizations and governments who ignore privacy as a basic human right.
It came stock on my OnePlus 13 as an integrated app.
I was able to run it as a guest, but then Google changed it so that you’re forced to log in with your Google account in order to use it.
Google literally has to ruin everything.
That’s exactly the “feature” they want to push. Nightmare scenario for users.
What’s the point of this? If AI is summarizing your email/texts, then AI writes a response, then AI on the other end does the same… aren’t we missing the purpose of communication between two people by having AI do it for us? I want to chat with, and email my family and friends, not an AI assistant.
All those years of sarcasm, shitposts, jokes, photos of your pets, and memes will be the reason you won’t gain entry into Nazi USA. 🤭
You don’t even have to be old. Death or serious illness/injury can affect us at any age, and it would suck if your family lost access to all the self-hosted photos and videos, for example.
“Make it usable” is a great idea.
Simplify as much as you can.
And remember, if you’re also self-hosting for family, someone will need to take over all that software and digital clutter when you’re gone.
I’ve been trimming as much as I can on my NAS, including only keeping the most important self-hosted software and heavily purging old files and backups.
The Meta AI app is a privacy disaster
So, it’s working as designed? 🤔
Why aren’t these devices, especially when they are sold/used in the public sector, not “private by design” with certain features locked down unless the user explicitly disables them?
It seems crazy to me that a plain HTTP stream from a hospital could be exposed to anyone remotely.
Bookmarking “headscale”!
I only recently started using Tailscale because it makes connecting to my local network through a Windows VM running in Boxes on Linux a hell of a lot easier than figuring out how to set up a networked bridge.
This sounds like a great alternative, and it looks like it can even work on a Synology NAS.
When you’re at work and using work devices, you have no reasonable expectation of privacy.
Trust between employers and employees can’t happen if one is spying on the other.
Privacy should be protected by law, especially voyeuristic shit like monitoring someone’s WFH camera or microphone.
Companies need to use better metrics for measuring performance that make sense, and very little of what they collect through this spyware makes sense.
If the effort and resources spent on spying were put into training, tools to help employees succed, and fair pay, these businesses would be much better off.
The reality is, they are using this monitoring tech to find any reason to terminate an employee without paying severance. Unethical means to an unethical end.
Common features include tracking time, keyboard strokes, email communications, websites visited, applications used and webcam video footage.
None of which are a measure of productivity or work completed.
Companies who employ spyware aren’t doing it for their business’s benefit.
Interesting, it’s slightly slower for me through the web interface both with a direct connect to my network, or when proxied through the internet. Still, we’re talking seconds here, and the results are so accurate!
Immich has effectively replaced the (expensive) Windows software Excire Foto, which I was using for on-device contextual search because Synology Photos search just sucks. Excire isn’t ideal to run from Linux because it has to be done through a VM, so I’m happy to self-host Immich and be able to use it even while out of the house.