What model to use for Smart Search, also its not working right now...
I think i want to use ViT-L-16-SigLIP2-256__webli
But whatever may be best for a rasp pi 5 a piece of server logs are attached
38 Replies
:wave: Hey @Defanoux,
Thanks for reaching out to us. Please carefully read this message and follow the recommended actions. This will help us be more effective in our support effort and leave more time for building Immich :immich:.
References
- Container Logs:
docker compose logs docs
- Container Status: docker ps -a docs
- Reverse Proxy: https://immich.app/docs/administration/reverse-proxy
- Code Formatting https://support.discord.com/hc/en-us/articles/210298617-Markdown-Text-101-Chat-Formatting-Bold-Italic-Underline#h_01GY0DAKGXDEHE263BCAYEGFJA
Checklist
I have...
1. :ballot_box_with_check: verified I'm on the latest release(note that mobile app releases may take some time).
2. :blue_square: read applicable release notes.
3. :blue_square: reviewed the FAQs for known issues.
4. :blue_square: reviewed Github for known issues.
5. :ballot_box_with_check: tried accessing Immich via local ip (without a custom reverse proxy).
6. :ballot_box_with_check: uploaded the relevant information (see below).
7. :blue_square: tried an incognito window, disabled extensions, cleared mobile app cache, logged out and back in, different browsers, etc. as applicable
(an item can be marked as "complete" by reacting with the appropriate number)
Information
In order to be able to effectively help you, we need you to provide clear information to show what the problem is. The exact details needed vary per case, but here is a list of things to consider:
- Your docker-compose.yml and .env files.
- Logs from all the containers and their status (see above).
- All the troubleshooting steps you've tried so far.
- Any recent changes you've made to Immich or your system.
- Details about your system (both software/OS and hardware).
- Details about your storage (filesystems, type of disks, output of commands like fdisk -l and df -h).
- The version of the Immich server, mobile app, and other relevant pieces.
- Any other information that you think might be relevant.
Please paste files and logs with proper code formatting, and especially avoid blurry screenshots.
Without the right information we can't work out what the problem is. Help us help you ;)
If this ticket can be closed you can use the /close command, and re-open it later if needed.
Latest version
Disk /dev/mmcblk0: 117.75 GiB, 126437294080 bytes, 246947840 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: dos Disk identifier: 0x9f88cf02 Device Boot Start End Sectors Size Id Type /dev/mmcblk0p1 8192 1056767 1048576 512M c W95 FAT32 (LBA) /dev/mmcblk0p2 1056768 246947839 245891072 117.3G 83 Linux Disk /dev/sda: 1.82 TiB, 2000398933504 bytes, 3907029167 sectors Disk model: Portable Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 4096 bytes I/O size (minimum/optimal): 4096 bytes / 4096 bytes Disklabel type: dos Disk identifier: 0x01444183 Device Boot Start End Sectors Size Id Type /dev/sda1 2048 3907026943 3907024896 1.8T 7 HPFS/NTFS/exFAT defanoux@raspberrypi:/mnt/nextcloud_data/immich-app $the model name is wrong, add the prefix of immich app
use this website to see which model you want
https://immich.app/docs/features/searching/#clip-models
then enter it like this
also please, use proper discord formatting and format your text a bit so it is less painful to read
You don't need that prefix. Just the model name.
How many gigs of ram does your pi 5 have? Please add the ML logs
without the prefix it does not work for me, then it just fails fetching it
In theory you just need the model name. What the documentation calls out and what I use
4
Ml logs ill brb
There's a lot more but its very repetitive
Which text
Can you remove the ML_CLIP_MODEL from the env, restart the stack. Then update the GUI with the model you want to use instead
actually where did you get those environment variables? I don't see where they exist
Which ones
The machine learning variables in the env file
Perhaps Gemini
Ill remove em
But still why is there so many errors in the logs
99% of my home lab is Gemini because a. Im lazy or b. Im not too skilled in this
You should read the actual docs instead :(
https://immich.app/docs/install/environment-variables/#machine-learning
As for why it's still failing, please try and copy and paste exactly the model from the documentation and save it, then try re-running the smart search job
ViT-L-16-SigLIP2-256__webli
So I chose this
I paste this in the model type
If you paste, save, and re-run smart search do you still get an error?
It seems to be running
I'm not sure if that will even run on your pi5. 4gb of RAM is really not a lot
Yeah I know
Its always at 66% util
It will probably take a long time to process as it may need to use swap because of not enough ram
Also make sure you aren't running off an SD card
Im using a 2tb hdd via USB
Is where my pics are at
Just make sure your data is backed up well according to 3-2-1 🙂
Is this alright
Mem going to burn ðŸ˜ðŸ˜
that model by itself uses 3 GB of RAM
you have 4 GB total
so I am not sure what you would expect
Maybe you would consider ViT-B-16-SigLIP__webli instead which uses 1.1 GB
I am currently trying
ViT-B-32__laion2b-s34b-b79k
Its like just over a gb
230% cpu util
Lmao 400
immich will use all the resources available to it, you need to set limits if you want it limited
Its okay just sometimes I cant send messages when it first started the processing
The rasp pi5 handled ViT-B-32laion2b-s34b-b79k pretty good
Is this better than ViT-B-32laion2b-s34b-b79k
This model works great