NEW 2vXpSwA7: anytest-v4 | openpose-v2_1 || abovzv: segment || bdsqlsz: canny | depth | lineart-anime | mlsdv2 | normal | normal-dsine | openpose | recolor | segment | segmentv2 | sketch | softedge | t2i-color-shuffle | tile-anime-α | tile-anime-β | tile-real || BRIA AI: bg-gen | canny | colorgrid | depth | NEW fill | openpose | recolor || CVL-Heidelberg: canny | depth || destitech: inpaint | inpaintv2 || diffusers: canny small | mid | full | depth small | mid | full | zoe || EcomXL: inpaint | softedge || Eugeoter: NEW noobai canny | depth | lineart-anime | lineart-real | mangaline | normal | scribble-pidi | scribble-hed | softedge-hed | tile | sdxl anime-canny | vidit-depth || h94: ip-adapter | ipa-vith | ipa+ | ipa+face || Hetaneko: pony canny | cannyv2 | color | depth | replicate | replicatev2 || HighCWu: canny-v3 || huchenlei: PuLID || Kataragi: canny | flatline | flatline-lora | inpaint | NEW ipa | line2color | line2color-lora | lineart | lineart-lora | NEW noob-ipa | recolor | recolor-lora | NEW rough-coating | tori29-blur | xdog-sketch || kohya-ss: real blur | canny | depth | anime blur | blur-beta | canny | depth | openpose | openposev2 | replicate | replicatev2 | scribble || PromeAI: lineart || ShermanG lineart || Stability.ai: canny | depth | recolor | revision | sketch || SargeZT: depth-16b-zoe | depth-faid-vidit | depth-zeed | depth-zoe | softedge | t2i-adapters binary | canny | color | depth | segmentation | softedge || TencentARC: canny | depth-midas | depth-zoe | lineart | openpose | recolor | sketch || TheMistoAI: mistoline | mistoline-lora || thibaud: openpose | openpose-lora || ttplanet: tile-real | tile-realv2 || NEW windsingai: pose | tile | tile-10w || xinsir: canny | cannyv2 | depth | openpose | openpose-twins | scribble | scribble-anime | tile | union | union-promax
ControlNetXL (CNXL) - A collection of Controlnet models for SDXL
(13.01.2025 - First NoobAI controlnets uploaded by Eugeoter)
(12.01.2025 - First Illustrious controlnets uploaded: windsingai-pose & -tile)
This collection strives to create a convenient download location of all currently available Controlnet models for SDXL. Please do read the version info for model specific instructions and further resources. If you are the author of one of these models and don't want it to appear here, please contact me to sort this out. The naming scheme of the files follows lllyasviel's from here https://huggingface.co/lllyasviel/sd_control_collection/tree/main.
CAUTION: The variants of controlnet models are marked as checkpoints only to make it possible to upload them all under one version, otherwise the already huge list would be even bigger. If you use downloading helpers the correct target folders are extensions/sd-webui-controlnet/models for automatic1111 and models/controlnet for forge/comfyui.
Requirements for Automatic1111: at least version 1.5.0, better upgrade to the latest version of 1.6.0. + sd-webui-controlnet extension v0.400+. Bugs or weird behaviour might occur. If you encounter any irregularities you can join us on our discord and ask for support or get in contact with the developers of Automatic1111/Controlnet via github/huggingface.
A huge thanks to all the authors, devs and contributors including but not limited to: abovzv, bdsqlsz, BRIA AI, CVL-Heidelberg, destitech, the diffusers institution, Eugeoter, h94, Hetaneko, HighCWu, huchenlei, lllyasviel, kataragi, kohya-ss, Mikubill, PromeAI, SargeZT, ShermanG, Stability.ai, TencentARC, TheMistoAI, thibaud, ttplanet, windsingai and xinsir.
Description
Recolor-Luminance model released by bdsqlsz on huggingface
FAQ
Comments (29)
What changed in new version?
Can anyone recommend a workflow such as MurphyLanga or Winstons, that has "controlnet lllite" nodes functionality built in and tested? for using models of ControlNet-LLLite form bdsqlsz
afaik they should be usable with every standard load controlnet model node there is? do you have any specific issues with getting them to work?
@eurotaku Some (and very significant for me) difficulties lie in adapting existing workwlows to the updated nodes of the controlnet-lllite; simply replacing one node will not work; i need to significantly rework the logic of worflows or (but most likely this is exactly what) create a work process from scratch. My skills won't allow me to do this in an acceptable amount of time, so I'm wondering if more experienced users have already done this
tried using this workflow:
https://github.com/kohya-ss/ControlNet-LLLite-ComfyUI
to get images like these:
https://huggingface.co/bdsqlsz/qinglong_controlnet-lllite
and nothing worked for me (the result with the node turned on and the node bypass was the same), it seems that the node "Load LLLite" has no effect... maybe I'm doing something wrong?
I realized what the problem was when using the test workspace - I incorrectly specified the value of the "end_percent" value. In most cases that I dealt with, this value was specified in the range from 0 to 1, but in the Load LLLite node this value should be specified in the range from 0 to 100, you need to get used to it. After I realized this and specified the correct value, everything worked as it should.
And yet, returning to my original question - there is a problem with integrating this node into existing workflows, and this is where I have big problems. In particular, I was unable to embed the node under discussion in my favorite and very convenient Winstons workflow https://civitai.com/models/138397
@modzhahead158 yeah, the end_percent thing is weird. but good to know, you won't be the only one getting confused by that :D
@eurotaku So, experimentally I found that the “Load LLLite” node conflicts with the “KSampler Inspire” node, which is used in the Winstons workflow, that I so dearly love. At this point my skills said astalavista, and I plunged into existential sadness. That's all for now )))
like this: https://i.imgur.com/VHoHiXx.png
@modzhahead158 is it not possible to disable the conflicting nodes (temporarily) if they are not in same workspace ?
@simartem07 You can disable nodes in the manager, is this what you need? as example: https://i.imgur.com/aM9N7Lk.png
@modzhahead158 thanks for the response but actually my reply was for your suffering in previous comments that you told some nodes are conflicting.. thats why i suggested you, not actually for me :)
@simartem07 sorry, I misunderstood you :) at that point the problem was already solved, but I will take your advice into consideration for the future
The new bdsqlsz models seem great! But—and please forgive me for seeming demanding or perhaps missing something—will FP16 versions of them possibly be made available? Since these versions don't allow for half precision, they are little challenging to use with limited vRAM. (Granted I'm on AMD so I know I have special issues.)
Thank you and bdsqlsz for your work and again sorry if this question/request is ignorant in some way.
absolutely no need to be sorry :)
i must admit that i am not entirely sure, if the fp16/32 thing even applies to those controlnet lllite (short for like lora lite) models. as far as i understood the approach they can be trained with different rank settings just like loras, stability.ai even provided two different rank versions of their models. but if someone here knows more about that, feel free to correct me :D
@eurotaku The error I got was "upsample_nearest2d_channels_last" not implemented for 'Half'" But this error does not appear when using my model of choice alone while using a half-precision instance of A1111. It only occurs when I try to apply the ControlNets while using half precision. I admit all of this is a bit above my pay grade.
@epzer hard to troubleshoot here, we provide help on our discord, though :)
i am still waiting for Tile support (not anime). Its been months since SDXL is released and still no tile :(.
Try blur instead.
@mweldonsd594 I did, not the same results at all.
only tile for anime still...
This XL controllnet model doesn't seem to work. For some reason, it has been upgraded to the latest version, but it still doesn't work
is there some specific error you're getting? you can join us on our discord for support, btw :)
maybe it's for ComfUI only?
@slartibart353 all these controlnets should work with any ui that supports controlnets for sdxl
@eurotaku Hi :-) as a newcomer - can u recommend an FAQ using the different preprocessors? for the ip-adapter i have three installed (ip-adapter_clip_sd15, ip-adapter_clip_sdxl, ip-adapter_clip_sdxl_plus_vith) - but when to use which? ;-) And maybe there are another out there doing a better work? :)
@slartibart353 you can start with controlnets in general using this guide https://stable-diffusion-art.com/controlnet/
to the second part: the first one is used with checkpoints based on sd1.5, the other two with sdxl ones. the plus indicates an enhanced variant, the exact differences are unknown to me, though. there are also face variants of those models, they essentially work the same but only use the face area of the provided image
