Ip adapter controlnet model github
Ip adapter controlnet model github. , ControlNet). pth] 2024-01-22 20:40:42,104 - ControlNet - INFO - ControlNet model ip-adapter-full-face_sd15 [3459c5eb] loaded. controlnet reference mode. 15, 2023. I was under the impression that the controlnet in forge was builtin and thought it would have code that differs compared to the regular extension. 5 / SDXL] Models [Note: need to rename model files to ip-adapter_plus_composition_sd15. The results look great with the original webui, but not as good with webui-forge. #2195. 9. The "locked" one preserves your model. Will upload the workflow to OpenArt soon. Modifying the pose vector layer to control character stances (Click for video) Mar 16, 2024 · An Image Prompt adapter (IP-adapter) is a ControlNet model that allows you to use an image as a prompt. reconstruction codes and make animatediff a diffusers plugin like sd-webui-animatediff. Enable Animatediff (any model, any animation type) Enable Controlnet (any model) Choose any image in tab (for processing by preprocessor) Run generation. co/models', make sure you don't have a local directory with the same name. 3, 2023. we present IP-Adapter, an effective and lightweight adapter to achieve image prompt capability for the pre-trained text-to-image diffusion models. controlnet conditioning scale - strength of controlnet. The model path is allowed to be longer though: you may place models in arbitrary subfolders and they will still be found. An IP-Adapter with only 22M parameters can achieve comparable or even better performance to a fine-tuned image prompt model. Jan 22, 2024 · For higher similarity, increase the weight of controlnet_conditioning_scale (IdentityNet) and ip_adapter_scale (Adapter). Successfully merging a pull request may close this issue. safetensors 2023-12-06 09:11:45,278 INFO Optional ControlNet model scribble for SD 1. Commit where the problem happens. 23, 2023. safetensors for sd-webui-controlnet extension to properly detect them. Add a color adapter (spatial palette), which has only 17M parameters. Sep 4, 2023 · If you were trying to load it from 'https://huggingface. Now you have to decide if you want to use IP-Adapter OR all the other controlnet models. 5 released #609. Development. IP-Adapter FaceID provides a way to extract only face features from an image and apply it to the generated image. Now we have ip-adapter-auto preprocessor that automatically pick the correct preprocessor for you. You can use it without any code changes. Enable IP-Adapter; Select ip-adapter_face_id as preprocessor and ip-adapter-faceid-portrait_sd15 as model; What should have happened? No milestone. guidance_scale - guidance scale value encourages the model to generate images closely linked to the text prompt at the expense of lower image quality. In other words, once IP-Adapter is trained, it can be directly reusable on custom models fine-tuned from the same base model. Jan 10, 2024 · Update 2024-01-24. Token merging is under construction now and the setting will not take effect. Any collaborations, suggestions or maybe a paid consultancy is highly appreciated. Using an IP-adapter model in AUTOMATIC1111. We have not supported multi-person yet, only use the largest face as reference facial landmarks. Jan 22, 2024 · yes, IP-Adapter plus ControlNet, i would also like to combine InstantID with other ip adapters 👍 1 shikasensei-dev reacted with thumbs up emoji All reactions Mar. safetensors & instant-ip-adapter. #2245 requests multiple images for single IP-Adapter unit because multiple IP-Adapter units do not seem to work together. Star 16k. It copys the weights of neural network blocks into a "locked" copy and a "trainable" copy. 9 💥 IP Adapter (Used for Increasing the Similarity of Batches of AnimateDiff Frames) 💥 Latent Consistency Model LoRA (LCM-LoRA) 💥 Latent Consistency Model (LCM) Native; 💥 Multi-ControlNet can be Combined with LCM, etc. 1. Step 2: Set up your txt2img settings and set up controlnet. The text was updated successfully, but these errors were encountered: Copy link. 2023-12-06 09:11:45,278 INFO Found ControlNet model inpaint for SD 1. Mar. The "regular" sdxl ip-adapter does work without issues. Want to ask about support for the updated IP Adapter XL model (ViT-h, plus version) and the IP Adapter face model. bin? I found only diffusion_pytorch_model. 431, IP Adapter does not work again Jan 19, 2024 As we freeze the original diffusion model in the training stage, the IP-Adapter can also be generalizable to the custom models fine-tuned from SD v1. anwoflow. However the new file is not supported out the box even when changing extension to pth, or using the provided safetensor file. 0 controlnet module:ip-adapter_clip_sdxl_plus_vith model: IPAdapter Composition [SD1. 5. join(model_dir, file_name)) if not os. Feb 10, 2024 · @lllyasviel I use the model Turbo Diffusion XL Turbo v1. py file can not recognize your safetensor files, some launchers from bilibili have already included the codes that @xiaohu2015 mentioned, but if you're using cloud services like autodl, you need to modify codes yourself, as those dockers are using the official controlnet scripts . 20. 5 Face ID Plus V2 as an example. It effectively acts like an 'instant LoRA' as @huchenlei Jan 22, 2024 · 2024-01-22 20:40:41,996 - ControlNet - INFO - Loaded state_dict from [D:\AINOVO\stable-diffusion-webui\models\ControlNet\ip-adapter-full-face_sd15. 12 with a CFG scale of 3. If not work, decrease controlnet_conditioning_scale. listdir('models') if os. Everything worked fine for months. Jan 23, 2024 · It should cause by ip-adapter_face_id preprocessor, because it will also happend when using other model like ip-adapter-faceid_sd15. When using the controlnetimg2img pipe, id_adapter is not work. See more info in the Adapter Zoo. At this point I think we are at the level of other solutions, but let's say we want the wolf to look just like the original image, for that I want to give the model more context of the wolf and where I want it to be so I'll use an IP adapter Jan 20, 2024 · SunGreen777 changed the title [Feature Request]: После обновления Controlnet 1. [Bug]: IP adapter controlnet fails #226 [Bug]: Control Net not working and throwing errors #259; Models, Checkpoints help! #267; I created a wiki page listing all known download sources. Otherwise, make sure 'models/image_encoder/' is the correct path to a directory containing a config. Topics customization personalization generative-art papers text-to-image image-synthesis deep-generative-model diffusion-models consistency-models text-to-video dall-e midjourney latent-diffusion stable-diffusion Nov 3, 2023 · The key is that your controlnet_model_guess. Drag and drop an image into controlnet, select IP-Adapter, and use the "ip-adapter-plus-face_sd15" file that you downloaded as the model. The easy fix that uncache IP-Adapter models has been landed. Reworking and adding content to an AI generated image. Mar 17, 2023 · Mikubill / sd-webui-controlnet Public. Maintainer. makedirs(model_dir, exist_ok=True) if not file_name: parts = urlparse(url) file_name = os. Commit where the Jan 22, 2024 · While I haven't tested it thoroughly, it seems like the portrait IP-adapter might be faster than others from the faceid family. ipynb Aug 23, 2023 · I change the controlnet demo from IPAdapter to IPAdapterPlus, while using "models/ip-adapter-plus_sd15. your implementation: original implementation: what was done in your implementation that's different? Introduction. Please feel free to add new items if I missed any. 430. Also i didn't get how to change style on local version, d Nov 5, 2023 · Is there an existing issue for this? I have searched the existing issues and checked the recent builds/commits of both this extension and the webui What happened? model: xl base 1. 5: control_v11p_sd15_inpaint_fp16. Feb 18, 2024 · 「IP-Adapter」とは、”Image Prompt Adapter”の略称であり、ControlNetの新しいモデルです。 これまではテキストプロンプトを用いて生成したい内容を入力していましたが、「IP-Adapter」を使うことで、画像自体がプロンプトの代わりとなって機能します。 Dec 22, 2023 · import gradio as gr import os import cv2 import numpy as np import torch from PIL import Image from insightface. 6. isdir(os Nov 10, 2023 · ip_adapter_sdxl_controlnet_demo: structural generation with image prompt. basename(parts. T2I-Adapters for SD 1. Fooocus Inpaint [SDXL] patch - Needs a little more Example Using FaceID V2XL but basically goes for every IPAdapter model First Picture with Endingstep 0. IP-Adapter/ip_adapter_demo. TODO. 00:52:17,542 - ControlNet - INFO - ControlNet model ip Jan 11, 2024 · I thought that the IP adapter FaceID should maybe behave like reference, at least when using an independant image because internally the preprocessor scan the image for a face. ip-adapter-plus-face_sd15. The "trainable" one learns your condition. 1 participant. (Edit: #98) Crop input image based on A1111 input checkbox removed. IP-Adapter can be generalized not only to other custom Mar 27, 2024 · I always prefer to allow the model to have a little freedom so it can adjust tiny details to make the image more coherent, so for this case I'll use 0. app import FaceAnalysis from diffusers import StableDiffusionPipeline, DDIMScheduler, AutoencoderKL from ip_adapter. from diffusers import StableDiffusionControlNetImg2ImgPipeline Feb 15, 2024 · I found that some users struggles to find download source of ControlNet models. In addition to controlnet, FooocusControl plans to continue to Introduction. is_available() is False. You switched accounts on another tab or window. load_ip_adapter(); However right now we do not support IPAdapterFull model checkpoints. As I understand it, IP adapter uses Clipvisiondetector which only supports CUDA or CPU. You should see following log line in your console: 2024-03-29 23:09:19,001 - ControlNet - INFO - ip-adapter-auto => ip-adapter_clip_g that indicates ip-adapter-auto is getting mapped to the actual preprocessor. I'm not sure if it was intentional or a typo, but the current line is different from the discussed one: sigma = sigma. 2s). The IP-Adapter is fully compatible with existing controllable tools, e. 431 опять не работает IP Adapter [Feature Request]: After updating Controlnet 1. 9s, load VAE: 0. What i already try: remove the venv; remove sd-webui-controlnet; Steps to reproduce the problem. . safetensors] PhotoMaker [SDXL] Original Project repo - Models. Made it possible to specify multiple ControlNetProcessors in pipeline's __call__() method (there is no limit to the number). cuda. For now i mostly found that Output block 6 is mostly for style and Input Block 3 mostly for Composition. Feb 14, 2023 · lllyasviel announced in Announcements. 2024-01-22 20:40:42,113 - ControlNet - INFO - Using preprocessor: ip-adapter_face_id_plus Feb 19, 2024 · when using the ip adapter-faceid-portrait-v11_sd15 model. 89 (I blacked out random NSFW stuff) Model Used :Juggernaut XLv9 Sampler : DPM++SDE karras 20 steps CFG 2 Jan 23, 2024 · It should cause by ip-adapter_face_id preprocessor, because it will also happend when using other model like ip-adapter-faceid_sd15. For higher text control ability, decrease ip_adapter_scale. ComfyUI's ControlNet Auxiliary Preprocessors. However, ip-adapter model also contains input image embedding, weight, and many values, so should not be cached. exists(cached_file): Oct 25, 2023 · IP Adapter XL + & IP Adapter Face #2195. The comparison of IP-Adapter_XL with Reimagine XL is shown as follows: Improvements in new version (2023. Reproduction. Where to download instantid-controlnet. 5 like other adapters (e. Dec 7, 2023 · However, I have an RX 5700XT video card and as far as I understand A1111 specifically limited its webui for RDNA1 so that pytorch is below version 2. pth. Steps to reproduce the problem. Command Line Arguments Apr 29, 2024 · Hey there, just wanted to ask if there is any kind of documentation about each different weight in the transformer index. update diffusers to 0. When you use ip adapter faceid in this repo, it follows the style and prompt of the model much more than the original implementation. Release T2I-Adapter. If you are a developer with your own unique controlnet model , with Fooocus-ControlNet-SDXL , you can easily integrate it into fooocus . 👍 9 dan4ik94, xiaohu2015, mityarko, gokuldaskumar, IAn2018cs, nicksonit, dancemanUK, el1xz, and LostXine reacted with thumbs up emoji. links. I loaded up a single IP adapter in controlnet with the intention of inpainting a face in img2img with the pre-processor "ip-adapter_face_id_plus" and model "ip-adapter-faceid-plusv2_sd15 [6e14fc1a]" ip_adapter_scale - strength of ip adapter. ip_adapter_faceid import IPAdapterFaceID # Function to list models in the 'models' folder def list_models(): return [f for f in os. solve/locate color degrade problem, check TDS_ solution, It seems that any color problems came from DDIM params. It's super important for IP Adapter "Instant LoRA" type workflows, and a whole lot faster than adding 2-4 different duplicate ControlNets to do the same thing. Also considering forge is a fork and as you say, some preprocessors do not work, then exclusion of them should not be handled on a main branch. Contribute to Fannovel16/comfyui_controlnet_aux development by creating an account on GitHub. bin" as adapter model checkpoint. lllyasviel. Mar 4, 2023 · Created a new ControlNetProcessor class and made it so that one is specified for each ControlNet processing. safetensors and ip-adapter. InstantID takes 2 models on the UI. Nov 15, 2023 · ip-adapter-full-face_sd15 - Standard face image prompt adapter. 7 participants. IP-Adapter/ip_adapter-full-face Feb 12, 2024 · Understanding Stable Diffusion, ControlNet, and IP-Adapter is crucial because it empowers users to tailor and mold the output, ensuring that the generated images align with the expectations and Nov 11, 2023 · IP Adapter Folks recently released a new IP face adapter shortly after safetensor support was added, less than 30 hours ago. dfaker started this conversation in General. The usage of other IP-adapters is similar. safetensors'] - Value not in list: dtype: '0. Add the depth adapter t2iadapter_depth_sd14v1. IP-Adapter can be generalized not only to other custom Mar 19, 2024 · Resolved SDXLClipModel issue 0Chencc/cog-face-to-many. (ipadapter model should be hooked first) Unit 0 Setting. You must set ip-adapter unit right before the ControlNet unit. The author of ipadapter has been contacted and is currently on the development schedule. What browsers do you use to access the UI ? No response. Chenlei Hu edited this page on Feb 15 · 9 revisions. We plan to train some models with "double controls", use two concat control maps and we are considering using images with holes as the second control map. safetensors - Plus face image prompt adapter. 👉 START FREE TRIAL 👈. safetensors and ip-adapter_plus_composition_sdxl. Feb 10, 2024 · Checklist The issue exists after disabling all extensions The issue exists on a clean installation of webui The issue is caused by an extension, but I believe it is caused by a bug in the webui The Bring back old Backgrounds! I finally found a workflow that does good 3440 x 1440 generations in a single go and was getting it working with IP-Adapter and realised I could recreate some of my favourite backgrounds from the past 20 years. Would it be possible to force CPU only just for IP Adapter model? RuntimeError: Attempting to deserialize object on a CUDA device but torch. 9 Second Picture with Endingstep 0. Jan 10, 2024 · IP-adapter controlnet img2img mat1 and mat2 shapes cannot be multiplied (2x1024 and 1280x768) #6516 Closed Honey-666 opened this issue Jan 10, 2024 · 4 comments Jan 24, 2024 · I tried to join IPA as an image control, but the following issue occurred and I am not sure if my usage is correct. 7000000000000001' not in ['fp16', 'fp32'] Output will be ignored Failed to validate prompt for output 1428: Output will be ignored If you're interested in understanding the diversity of use cases for foundation model adaptation (potentially beyond the specific adapters supported by Refiners), we suggest you take a look at these outstanding papers: ControlNet; T2I-Adapter; IP-Adapter; Medical SAM Adapter; 3DSAM-adapter; SAM-adapter; Cross Modality Attention Adapter; UniAdapter The main model can be downloaded from HuggingFace and should be placed into the ComfyUI/models/instantid directory. Oct 29, 2023 · Fooocus-ControlNet-SDXL simplifies the way fooocus integrates with controlnet by simply defining pre-processing and adding configuration files. 0 because it doesn't work with my video card model. You only need to follow the table above and select the appropriate preprocessor and model. ip-adapter_sdxl is working. You signed out in another tab or window. Step 1: Select a checkpoint model The full prompt is below if you're curious. This page documents multiple sources of models for the integrated ControlNet extension. 8): Switch to CLIP-ViT-H: we trained the new IP-Adapter with OpenCLIP-ViT-H-14 instead of OpenCLIP-ViT-bigG-14. """ os. (Note that the model is called ip_adapter as it is based on the IPAdapter). Please follow the guide to try this new feature. Also, its ability to blend faces effectively and maintain consistency across various prompts and seeds is quite remarkable (have a look at the image below). For specific styles, choose corresponding base model makes differences. Describe the bug. IP Adapter XL + & IP Adapter Face. , ControlNet and T2I-Adapter. If there are multiple matches, any files placed inside a krita subfolder are prioritized. Oct 7, 2023 · ip-adapter_sdxl_vit-h / ip-adapter-plus_sdxl_vit-h are not working. Notifications. Important: set your "starting control step" to about 0. You should always set the ipadapter model as first model, as the ControlNet model takes the output from the ipadapter model. support IP-Adapter. Adding detail and iteratively refining small parts of the image. Sep 25, 2023 · Saved searches Use saved searches to filter your results more quickly Jan 2, 2024 · Steps to reproduce the problem. Although ViT-bigG is much larger than ViT-H, our The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt. Feb 11, 2023 · ControlNet is a neural network structure to control diffusion models by adding extra conditions. Made it so that controlnet is not specified in the Pipeline constructor. Jan 19, 2024 · Now, I have to engineer that how to combine your training script with that of controlnet script. Nov 2, 2023 · with both ip-adapter-plus_sdxl_vit-h and ip-adapter-plus-face_sdxl_vit-h (using safetensor files), with both the SDXL clip preprocessor and the vith preprocessor, and they don't seem to generate expected results. Add four new adapters style, color, openpose and canny. You also needs a controlnet, place it in the ComfyUI controlnet directory. g. Installation of insightface. 6s, forge load real models: 0. This will lead to some model like "depth-aware inpainting" or "canny-edge-aware inpainting". path. No branches or pull requests. Image-guided image-to-image and inpainting can be also achieved by simply replacing text prompt with Mar 7, 2024 · 2024-03-06 18:38:17,907 - ControlNet - INFO - IP-Adapter faceid plus v2 detected. 5 not found (search path: control_v11p_sd15_scribble, control_lora_rank128_v11p_sd15_scribble) 2023-12-06 09:11:45,278 INFO Found ControlNet model scribble Yes, please! I was so stoked to see it added, and bummed to see it's now gone again. Launch a generation with ip-adapter_sdxl_vit-h or ip-adapter-plus_sdxl_vit-h. Using ControlNet to guide image generation with a crude scribble. Sep 10, 2023 · """Download a file from url into model_dir, using the file present if possible. Fork 1. 2024-02-06 13:42:33,775 - ControlNet - INFO - ControlNet Input Mode: InputMode. 2024-03-06 18:38:18,176 - ControlNet - INFO - Using preprocessor: ip-adapter_face_id_plus You signed in with another tab or window. 2024-03-06 18:38:18,172 - ControlNet - INFO - ControlNet model ip-adapter-faceid-plusv2_sd15 6e14fc1a loaded. Sep 18, 2023 · I have referred to the following comments. 5 days ago · You need to select the ControlNet extension to use the model. It only happens for me if i type in the manual weights to fast. path) cached_file = os. controlnet from TDS4874. anyone interested in adding it Oct 15, 2023 · I've tried to use the IP Adapter Controlnet Model with this port of the WebUI but it failed. bin and the results are not that good. raisindetre commented last week A current workaround that has worked for me - need to do more testing to see if it's consistent - is to add the command-line flag "--no-half" to your launch settings. ip_model = IPAdapterPlus(pipe, image_encoder_path, ip_ckpt, device) The ControlNet Update shouldnt have changed how IP-Adapter was working. Face swap via diffusion models [Lora+IP-Adapter+Controlnet+text embedding optimization] - somuchtome/Faceswap Feb 15, 2024 · ControlNet model download. Use the DirectML Fork of Automatic1111 (without using --onnx or olive) Download the IP-Adapter Models for ControlNet Begin to load 1 model Model loaded in 1. Our method not only outperforms other methods in terms of image quality, but also produces images that better align with the reference image. Image preprocessing was also moved here. on Feb 12, 2023. The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt. Beta Was this translation helpful? Jan 28, 2024 · ipadapter model; ControlNet model; How to use. After downloading the models, move them to your ControlNet models folder. What should have happened? Using the control model. a good start is using openpose controlnet (but replace condition with image condition) + IP-Adapter (with image condition) Can't find a way to get ControlNet preprocessor: ip-adapter_face_id_plus And, sorry, no, InsightFace+CLIP-H produces way different images compared to what I get on a1111 with ip-adapter_face_id_plus even using the same model. 9s (load weights from disk: 0. SIMPLE Model paths must contain one of the search patterns entirely to match. Reproduction: Following 2 setups use exact IP-Adapter Nov 22, 2023 · we recently added IP-adapter support to many of our pipelines including all the text2img, img2img and inpaint pipelines, as well as the text2img ControlNet pipeline. Read the article IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models by He Ye and coworkers and visit their Github page for implementation details. abspath(os. Feb 6, 2024 · The recent changes to the Ip-adapter code still freeze Directml. For over-saturation, decrease the ip_adapter_scale. Papers and resources on Controllable Generation using Diffusion Models, including ControlNet, DreamBooth, T2I-Adapter, IP-Adapter. This new face adapter appears to be significantly better at stylizing faces and I would like it ip_adapter_scale：IPA 强度; controlnet_conditioning_scale：ID Controlnet 强度; step：步数，官方默认30步; guidance_scale：提示词相关度，一般默认为5; enhance_face_region：脸部增强选项 🆕; seed：种子 Inpainting on a photo using a realistic model. SDXL FaceID Plus v2 is added to the models list. json file. Returns the path to the downloaded file. We recently added IP-adapter support to many of our pipelines in diffusers! You can now very easily load your IP-Adapter into a diffusers pipeline with pipe. Nov 1, 2023 · IP-Adapter does not work in controlnet, There are no errors, the effect is as if controlnet is not enabled. but failed in loading ip-adapter. IP-Adapter/ip_adapter-plus-face This includes using multiple inputs or the batch directory option within controlnet for IPadapter. Apr 21, 2024 · You need to rename model files to ip-adapter_plus_composition_sd15. InstantID [SDXL] Original Project repo - Follow instruction in here. Remember at the moment this is only for SDXL. I will use the SD 1. anwoflow started this conversation in General. Stable Diffusion in the Cloud⚡️ Run Automatic1111 in your browser in under 90 seconds. just solved it, works fine. Somehow the recommendation of fonik000 worked to show the exact same options and preprocessors that the original CN has, but there were some errors here and there, so I decided to go back to the integrated CN, and to be honest after testing I see that the pre-installed preprocessor in this integrated CN "InsightFace+CLIP-H (IPAdapter)" does pretty good - if not better - job than the "ip Dec 11, 2023 · For over-saturation, decrease the ip_adapter_scale. You signed in with another tab or window. dfaker. item() if sigma is not None else 999999999. Reload to refresh your session. Enable IP-Adapter; Select ip-adapter_face_id as preprocessor and ip-adapter-faceid-portrait_sd15 as model; What should have happened? Dec 24, 2023 · This model is not currently supported because it uses the external insightface library. If this happens it actually dont register the Inputs and the Output picture is the same as if you didnt manually change the weight. Feb. Apr 22, 2024 · If you manually type in the Weights for the ne Weight Type in IP Adapter XL to fast Controlnet doesn't get the changes and the Chart disappears. on Oct 25, 2023. 1s, calculate empty prompt: 0. webui: 1. Jan 15, 2024 · See #2469, multiple IP-Adapter was broken because the original design of model caching mechanism assumes that model file name + model hash uniquely maps to a model object, but IP-Adapter model contains more information and cannot be uniquely idenfitied by file name + hash. Nov 21, 2023 · yiyixuxu commented on Nov 21, 2023. code: ipa_model_path = f'/ipa_models' device = "cuda" face_adapter = f'InstantIDm * IPAdapterUnifiedLoader 1431: - Exception when validating inner node: tuple index out of range * IPAdapter 1430: - Required input is missing: clip_vision - Value not in list: model_name: '0' not in ['ioclab_sd15_recolor. 9k. 💥 Prompt Weighting and Long Prompts (Compel) 💥 DreamBooth & LoRA; 💥 FFMPEG Interpolation The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt. 0 controlnet: 1. bin ignores the pose from ControlNet OpenPose, do I understand correctly that ControlNet does not work with the model? The text was updated successfully, but these errors were encountered: 2024-04-01 00:04:12,215 - ControlNet - INFO - ip-adapter-auto = > ip-adapter_face_id_plus 2024-04-01 00:04:12,215 - ControlNet - INFO - unit_separate = False, style_align = False 2024-04-01 00:04:12,510 - ControlNet - INFO - Loading model: ip-adapter-faceid-plusv2_sdxl [187cb962] 2024-04-01 00:04:12,774 - ControlNet - INFO - Loaded state_dict You signed in with another tab or window. yk xp du rt ba pg yk og uj va
1