[DeviantArt] Site Change - DeviantArt Now Limits Downloads to 10 Per Week

[sourmilk01]

Encountered an issue downloaded from DeviantArt today, and after looking into it its looks like the website has implemented a major change:

https://www.deviantart.com/team/status-update/An-adjustments-being-made-to-1307747979

Downloads for free accounts (non-CORE subscribing accounts) are now limited to 10 downloads per week.

Downloads for CORE subscribing accounts are limited to only 150 downloads per week.

This looks like it might be the end of scraping source-quality files from DeviantArt, unless they revert the changes or if you can schedule out 150 downloads a week with a paying subscription. If you'd like, you can leave a comment on the thread above and voice your displeasure in the comments.

Not really sure what to think of this, @mikf do you have any feedback?

[GiovanH]

This appears to be manifesting as an unidentified 500 Internal Server Error:

[urllib3.connectionpool][debug] Starting new HTTPS connection (1): www.deviantart.com:443
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /api/v1/oauth2/user/profile/robbyrackleff HTTP/1.1" 200 362
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /robbyrackleff/art/118485205 HTTP/1.1" 200 None
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /api/v1/oauth2/deviation/6D9435B6-239A-A81F-8AB1-31502E219A25 HTTP/1.1" 200 1207
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /api/v1/oauth2/deviation/metadata?deviationids%5B0%5D=6D9435B6-239A-A81F-8AB1-31502E219A25&mature_content=true HTTP/1.1" 200 367
[deviantart][debug] Active postprocessor modules: [MetadataPP]
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /api/v1/oauth2/deviation/download/6D9435B6-239A-A81F-8AB1-31502E219A25?mature_content=true HTTP/1.1" 500 83
[deviantart][debug] '500 Internal Server Error' for 'https://www.deviantart.com/api/v1/oauth2/deviation/download/6D9435B6-239A-A81F-8AB1-31502E219A25?mature_content=true' (1/5)
[deviantart][debug] Sleeping 1.00 seconds (retry)

[StickyChannel92]

[deviantart][error] API responded with 400 Bad Request
This is what appears for me. For me, the images still do download, but the queue takes longer, and also affects downloads via the main account.

[GiovanH]

There may be something else failing as a result of this change as part of my 500s, as I don't think I've hit 40 images this month. This does seem linked to yesterday's change.

[sourmilk01]

For context, my error manifests as:

[deviantart][debug] Sleeping 5.00 seconds (api)
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /api/v1/oauth2/deviation/download/191F40FD-90A3-17A1-B802-6AB0268F38DD?mature_content=true HTTP/1.1" 400 107
[deviantart][debug] {"error":"invalid_request","error_description":"Free download limit reached.","error_code":2,"status":"error"}
[deviantart][error] An unexpected error occurred: KeyError - 'src'. Please run gallery-dl again with the --verbose flag, copy its output and report this issue on https://github.com/mikf/gallery-dl/issues .
[deviantart][debug]
Traceback (most recent call last):
  File "gallery_dl\job.py", line 163, in run
  File "gallery_dl\job.py", line 222, in dispatch
  File "gallery_dl\extractor\deviantart.py", line 198, in items
  File "gallery_dl\extractor\deviantart.py", line 313, in commit
KeyError: 'src'

Also, trying to manually download from a DA post that allows it results in: "Download limit reached
You've used all of your 10 downloads for this week. Upgrade to Core to unlock more weekly downloads."

[jamarcka-creator]

It seems download is possible pressing F12 and going to Network - Media... but it's a pain. I guess that if the video is still downloaded to the browser, then technically gallery-dl should be able to get it, but I'm still getting the error reported at the top.

[tp0]

It seems download is possible pressing F12 and going to Network - Media... but it's a pain. I guess that if the video is still downloaded to the browser, then technically gallery-dl should be able to get it, but I'm still getting the error reported at the top.

Plain URL to the lower quality image / mp4 file is right on the page source inside img / video src tags. URL itself has no auth key and not associated to cookies, but mature-marked media URL only shows up when logged in and the higher quality version is behind the download button.

[foducool]

deviantart really has a knack for shooting themselves in the foot, love that even paying subscribers are impacted with this, this is definitely gonna end well lmao
so, from what I gathered, it only happens on deviations that provides a "free download link", which probably appears at the poster's discretions through this option :

Seems like the extractor is made in such a way that if any kind of download link is provided, then gallery-dl will favor downloading the deviation through it instead of doing it the hard way, which then counts towards the download limit
I get this error when it happens btw :
[deviantart][debug] {"error":"invalid_request","error_description":"Free download limit reached.","error_code":2,"status":"error"}

[geoffk777]

I'm getting this error too with OAauth on DeviantArt. Some creators work fine but others, including ones that I'm subscribed too bomb with a similar error:

PS C:\Gallery-dl> .\gallery-dl --verbose -d "f:" "https://www.deviantart.com/BigJuggsAI/gallery/all"
[gallery-dl][debug] Version 1.31.5 - Executable (stable/windows)
[gallery-dl][debug] Python 3.13.7 - Windows-10-10.0.19043-SP0
[gallery-dl][debug] requests 2.32.5 - urllib3 2.6.3
[gallery-dl][debug] Configuration Files ['%USERPROFILE%\gallery-dl.conf']
[gallery-dl][debug] Starting DownloadJob for 'https://www.deviantart.com/BigJuggsAI/gallery/all'
[deviantart][debug] Using DeviantartGalleryExtractor for 'https://www.deviantart.com/BigJuggsAI/gallery/all'
[deviantart][debug] Using custom API credentials (client-id 62147)
[deviantart][debug] Sleeping 2.00 seconds (api)
[urllib3.connectionpool][debug] Starting new HTTPS connection (1): www.deviantart.com:443
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /api/v1/oauth2/user/profile/bigjuggsai HTTP/1.1" 200 1316
[deviantart][debug] Sleeping 2.00 seconds (api)
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /api/v1/oauth2/gallery/all?username=BigJuggsAI&offset=0&limit=24&mature_content=true HTTP/1.1" 200 7871
[deviantart][debug] Sleeping 2.00 seconds (api)
[urllib3.connectionpool][debug] https://www.deviantart.com:443 "GET /api/v1/oauth2/deviation/download/17617D1C-FC25-88CC-0128-0C81A416F29E?mature_content=true HTTP/1.1" 400 107
[deviantart][debug] {"error":"invalid_request","error_description":"Free download limit reached.","error_code":2,"status":"error"}

But it's not every page--just some of them. So it's not just a simple limit. In fact, with no refresh token, the same page will download a couple of free files. But with the refresh token cached, it bombs immediately as above.

[foducool]

@geoffk777 yeah as I said, it's because gallery-dl is trying to use the download link to download the deviation and they recently capped this at 10 per week for normal users. it isn't tied to the deviation being paid content or not (you can raise the limit to 150 if you pay for core but screw 'em for putting a limit even on those lmao)
those that are going through normally must be because the download option wasn't enabled for this deviation (problem is that if you it the limit and a post with a download button, that'll block everything after it from downloading, can't even try to bypass it by using the range option)
for now I'm finding ways to copy the links of the whole list of deviations I want through a clever use of the browser's html inspector and notepad++ then keep 'em in a fil and doing gallery-dl -i with it
that won't download those that have download links but at least you'll be able to download there rest (and go download those with download links manually if you really want them)

[mikf]

It is possible to disable limited original downloads entirely and always use the preview versions by setting original to false: -o original=false

[foducool]

setting original to false didn't work on my end, maybe because it's a video?

[Vetches]

Briefly chiming in to add that I also tried -o original=false and found it to work for images but not videos.

[geoffk777]

Thanks for the explanation. original=false didn't seem to work for me, either on the command line or in the conf file (where it was set to True).

I don't know how deviantart expects users to download creators with 10,000 plus images and videos. Even with a core membership, that would take a year or more at 150 a week, which is ridiculous.