- Update ComfyUI
- Add Kolors native support
- Add Kolors IPAdapter support
Weights:
- Update ComfyUI-Impact-Pack
New controlnets:
- controlnet-depth-sdxl-1.0.safetensors
- controlnet-union-sdxl-1.0.safetensors
- TTPLANET_Controlnet_Tile_realistic_v2_rank256.safetensors
New checkpoints:
- Stable Cascade (checkpoint, unet and vae)
- PixArt-Sigma-XL-2-1024-MS.pth
Add more ultralytics detectors:
- bbox/face_yolov8n.pt
- bbox/face_yolov8n.pt
- bbox/face_yolov8n_v2.pt
- bbox/face_yolov8s.pt
- bbox/hand_yolov8n.pt
- segm/deepfashion2_yolov8s-seg.pt
- segm/face_yolov8m-seg_60.pt
- segm/face_yolov8n-seg2_60.pt
- segm/hair_yolov8n-seg_60.pt
- segm/hair_yolov8n-seg_60.pt
- segm/person_yolov8n-seg.pt
- segm/person_yolov8s-seg.pt
- segm/skin_yolov8m-seg_400.pt
- segm/skin_yolov8n-seg_400.pt
- segm/skin_yolov8n-seg_800.pt
Fixes:
- Allow duplicate URLs within workflow (fixes #120)
New checkpoints:
New loras:
- hd_helper_v1.safetensors an SD15 helper for larger images
- Polaroidv2.safetensors (SDXL)
- Update ComfyUI
- Add ComfyMath custom nodes
- Add SUPIR custom nodes
- Add KJ's IC Light custom nodes
- Add StyleAligned custom nodes
- Add AttentionCouple custom nodes
- TensorRT engines added for SD3 on A40, A100 and H100
- ComfyUI enums helper added for easy access to schedulers and samplers
- Update ReActor to latest
New weights:
- aura_flow_0.1.safetensors
- Kolors diffusers model and LLMs:
- chatglm3-4bit.safetensors
- chatglm3-8bit.safetensors
- chatglm3-fp16.safetensors
SUPIR weights:
- SUPIR-v0F.ckpt
- SUPIR-v0F_fp16.safetensors
- SUPIR-v0Q.ckpt
- SUPIR-v0Q_fp16.safetensors
LivePortrait weights:
- appearance_feature_extractor.safetensors
- landmark.onnx
- motion_extractor.safetensors
- spade_generator.safetensors
- stitching_retargeting_module.safetensors
- warping_module.safetensors
FaceRestore weights:
- GPEN-BFR-1024.onnx
- GPEN-BFR-2048.onnx
- GPEN-BFR-512.onnx
Lora:
- Update ComfyUI
- Update ComfyUI_IPAdapter_plus
- Update ComfyUI-AnimateDiff-Evolved
- Update ComfyUI-Advanced-ControlNet (not backwards compatible)
- Update ComfyUI-VideoHelperSuite
- Update comfyui-prompter-fofrai
New weights:
- ponyDiffusionV6XL_v6StartWithThisOne.safetensors
- SD3-Controlnet-Canny.safetensors
- SD3-Controlnet-Pose.safetensors
- SD3-Controlnet-Tile.safetensors
- controlnet-tile-sdxl-1.0.safetensors
- Add comfyui-prompter-fofrai custom node
Visit the train tab on Replicate to create a new version of this model with your own weights:
https://replicate.com/fofr/any-comfyui-workflow/train
Here you can give public or private URLs to weights on HuggingFace and CivitAI. If URLs are private or need authentication, make sure to include an API key or access token.
Check the training logs to see what filenames to use in your workflow JSON. For example:
Downloading from HuggingFace:
...
Size of the tar file: 217.88 MB
====================================
When using your new model, use these filenames in your JSON workflow:
araminta_k_midsommar_cartoon.safetensors
Add SD3 weights and CLIP weights:
- sd3_medium.safetensors
- sd3_medium_incl_clips.safetensors
- sd3_medium_incl_clips_t5xxlfp16.safetensors
- sd3_medium_incl_clips_t5xxlfp8.safetensors
- clip_g.safetensors
- clip_l.safetensors
- t5xxl_fp16.safetensors
Updated ComfyUI to latest. Includes SD3 nodes.
Updated ComfyUI custom nodes:
- ComfyUI_IPAdapter_plus
- ComfyUI-Inspire-Pack
- comfyui-tooling-nodes
- ComfyUI-AnimateDiff-Evolved
- ComfyUI-VideoHelperSuite
- ComfyUI-Advanced-ControlNet
- efficiency-nodes-comfyui
- Derfuu_ComfyUI_ModdedNodes
- ComfyUI_FizzNodes
- ComfyUI_tinyterraNodes
- ComfyUI-Custom-Scripts
- masquerade-nodes-comfyui
- comfyui-reactor-node
- was-node-suite-comfyui
- comfyui_segment_anything
- ComfyUI-BRIA_AI-RMBG
- ComfyUI-layerdiffuse
- ComfyUI-KJNodes
- PuLID_ComfyUI
- comfyui_jankhidiffusion
- ComfyUI-Anyline
- rgthree-comfy
New checkpoints:
Add araminta_k’s new loras, most use "daiton style" as keyword:
- araminta_k/abstract-anime-painting.safetensors,
- araminta_k/araminta-k-illustration.safetensors,
- araminta_k/BandW-Manga.safetensors,
- araminta_k/gemini-anime.safetensors,
- araminta_k/haute-couture-illustration.safetensors,
- araminta_k/japanese-poster-prints.safetensors,
- araminta_k/lost-and-lonely-manga.safetensors,
- araminta_k/painting-light.safetensors,
- araminta_k/pop-art-anime.safetensors,
- araminta_k/sketched-out-manga.safetensors,
- araminta_k/soft-ones.safetensors,
- araminta_k/soft-pastel-anime.safetensors,
- araminta_k/vincent-sketch-manga.safetensors
Other new loras:
- RW_PS1v1.safetensors (PS1 SD15 lora, "ps1 style")
- artificialguybr/3DRedmond15V-LiberteRedmond-3DRenderStyle-3DRenderAF.safetensors ("3D Render Style, 3DRenderAF")
New checkpoints:
New embedding:
Models for ToonCrafter:
- CLIP-ViT-H-fp16.safetensors
- stable-diffusion-2-1-clip-fp16.safetensors
- stable-diffusion-2-1-clip.safetensors
- tooncrafter_512_interp-fp16.safetensors
New detailer:
Update KJNodes to latest:
- Refactor repo to make it easier to use as a template
- Added guide in
MAKING_A_MODEL_GUIDE.md
New SDXL control nets:
- controlnet-openpose-sdxl-1.0.safetensors
- controlnet-openpose-sdxl-1.0_twins.safetensors
- controlnet-scribble-sdxl-1.0.safetensors
- controlnet-canny-sdxl-1.0.safetensors
- controlnet-canny-sdxl-1.0_V2.safetensors
Update comfyui_controlnet_aux to include support for:
- Anyline
- Metric3D
- TTPlanet Tile
Update ComfyUI-AnimateDiff-Evolved to latest and add support for:
- CameraCtrl (CameraCtrl_pruned.safetensors)
Update custom nodes:
- ComfyUI_InstantID to fix PuLID compatibility
- ComfyUI-Frame-Interpolation
- PuLID_ComfyUI
- ComfyUI-VideoHelperSuite
- ComfyUI-Advanced-ControlNet
- efficiency-nodes-comfyui
- ComfyUI_FizzNodes
- ComfyUI_UltimateSDUpscale
- ComfyUI_essentials
- ComfyUI-Inspire-Pack
Add support for GLIGEN weights:
- gligen_sd14_textbox_pruned.safetensors
- gligen_sd14_textbox_pruned_fp16.safetensors
Add support for custom nodes:
New models:
- crystalClearXL_ccxl.safetensors
- Anyline preprocessor MTEED.pth
Add support for HiDiffusion via comfyui_jankhidiffusion
Add support for ComfyUI-IC-Light-Native
New UNets for IC Light:
- iclight_sd15_fbc_unet_ldm.safetensors
- iclight_sd15_fc_unet_ldm.safetensors
New checkpoints:
Added MistoLine lineart controlnets:
- mistoLine_fp16.safetensors
- mistoLine_rank256.safetensors
Add PuLID-ComfyUI custom node support
New checkpoints:
New checkpoints:
- Adding support for rembg, used in a variety of custom nodes
- Update ComfyUI_tinyterraNodes to latest to fix rembg issue
Update ComfyUI Controlnet Aux
New controlnets:
New controlnet preprocessors:
- Diffusion-Edge (Paper)
- Depth-Anything and Zoe Depth-Anything
- Unimatch optical flow estimation
- TEED
- Update ComfyUI to latest (eecd69b)
New checkpoint:
Add Hyper-SD Lora and UNet weights:
- Hyper-SD15-1step-lora.safetensors
- Hyper-SD15-2steps-lora.safetensors
- Hyper-SD15-4steps-lora.safetensors
- Hyper-SD15-8steps-lora.safetensors
- Hyper-SDXL-1step-lora.safetensors
- Hyper-SDXL-2steps-lora.safetensors
- Hyper-SDXL-4steps-lora.safetensors
- Hyper-SDXL-8steps-lora.safetensors
- Hyper-SDXL-1step-Unet-Comfyui.fp16.safetensors
- Hyper-SDXL-1step-Unet.safetensors
Update ComfyUI_Essentials to latest as requested in #71
New SD15 loras and embeddings as requested in #72:
- age_slider-LECO-v1.safetensors
- weight_slider-LECO-v1.safetensors
- SDXLrender_v2.0.safetensors
- JuggernautNegative-neg.pt
Also added ultralytics model hair_yolov8n-seg_60.pt
New AnimateDiff weights for AnimateLCM:
- AnimateLCM_sd15_t2v.ckpt
- AnimateLCM_sd15_t2v_lora.safetensors
New checkpoint:
- Add ComfyUI-layerdiffuse custom node
- layer_sd15_bg2fg.safetensors
- layer_sd15_fg2bg.safetensors
- layer_sd15_joint.safetensors
- layer_sd15_transparent_attn.safetensors
- layer_sd15_vae_transparent_decoder.safetensors
- layer_xl_bg2ble.safetensors
- layer_xl_bgble2fg.safetensors
- layer_xl_fg2ble.safetensors
- layer_xl_fgble2bg.safetensors
- layer_xl_transparent_attn.safetensors
- layer_xl_transparent_conv.safetensors
- vae_transparent_decoder.safetensors
Update dependencies:
- IPAdapter plus to support new adapter
ip-adapter-faceid-portrait_sdxl_unnorm.binfor "extreme profile styles" - ComfyUI
- AnimateDiff-Evolved
- VideoHelperSuite
- Advanced-ControlNet
- Update the Impact-Pack custom node to support latest IPAdapter-Plus nodes.
- Add support for comfyui-tooling-nodes to add base64 image handling
- Add option to return WebP optimised images from model
- Fix PhotoMaker by using official implementation and updating custom node
AnimateDiff LCM Motion module:
New models added:
- rundiffusionXL_beta.safetensors
- pixlAnimeCartoonComic_v10.safetensors
- realisticLCMBYStable_v10.safetensors
- photonLCM_v10.safetensors
New loras:
- pk_trainer_xl_v1.safetensors
- glowneon_xl_v1.safetensors
- add-detail-xl.safetensors
- aesthetic_anime_v1s.safetensors
New embeddings:
- Update ComfyUI to latest (6c6a392)
- Update to latest IPAdapter Plus ( 1ac1cae) (breaking changes, custom nodes re-written)
New SD1.5 checkpoints for AnimateDiff:
- dynavision_v20Bakedvae.safetensors
- imp_v10.safetensors
- magicmixReverie_v10.safetensors
- majicmixRealistic_v7.safetensors
- rcnzCartoon3d_v20.safetensors
- toonyou_beta6.safetensors
Add ip_plus_composition_sdxl.safetensors IPAdapter
Add AnimateDiff lightning weights:
- animatediff_lightning_1step_comfyui.safetensors
- animatediff_lightning_1step_diffusers.safetensors
- animatediff_lightning_2step_comfyui.safetensors
- animatediff_lightning_2step_diffusers.safetensors
- animatediff_lightning_4step_comfyui.safetensors
- animatediff_lightning_4step_diffusers.safetensors
- animatediff_lightning_8step_comfyui.safetensors
- animatediff_lightning_8step_diffusers.safetensors
Add ip_plus_composition_sd15.safetensors IPAdapter
Add Juggernaut v9 and lightning:
- Juggernaut-XL_v9_RunDiffusionPhoto_v2.safetensors
- Juggernaut_RunDiffusionPhoto2_Lightning_4Steps.safetensors
Add Albedobase XL v2.1:
- albedobaseXL_v21.safetensors
Add QR Monster control nets for illusions:
- control_v1p_sd15_qrcode_monster.safetensors
- control_v1p_sdxl_qrcode_monster.safetensors
Add RealVis XL V4 Lightning:
- RealVisXL_V4.0_Lightning.safetensors
- Fix: Always cancel and clear the ComfyUI server queue after a prediction
- Add support for ComfyUI_Comfyroll_CustomNodes
- Correctly download embeddings nested within prompt strings
- Add safetensor SDXL lightning checkpoints, UNETs and Loras
- Add SDXL lightning UNETs and Lora weights
Models added:
- sdxl_lightning_2step_lora.pth
- sdxl_lightning_4step_lora.pth
- sdxl_lightning_8step_lora.pth
- sdxl_lightning_1step_unet_x0.pth
- sdxl_lightning_2step_unet.pth
- sdxl_lightning_4step_unet.pth
- sdxl_lightning_8step_unet.pth
- epicrealismXL_v10.safetensors
- RealVisXL_V4.0.safetensors
- Add support for WAS nodes
- Add support for comfyui-segment-anything
Models added:
- mobile_sam.pt
- sam_hq_vit_b.pth
- sam_hq_vit_h.pth
- sam_hq_vit_l.pth
- groundingdino_swinb_cogcoor.pth
- groundingdino_swint_ogc.pth
- absolutereality_v181.safetensors
- PerfectEyesXL Lora
- More details Lora
- FastNegative embedding
- Add support for ComfyUI-Impact-Pack
- Add ViT-H, ViT-L and ViT-B segment anything models
- Update ComfyUI-AnimateDiff-Evolved to support Gen2 nodes (Fixes #19)
- Add support for InstantID
- Add support for comfyui-reactor-nodes
- Improve support for Insightface in ComfyUI_IPAdapter_plus
- Fix FaceID lora location
Models added:
- instantid-ipadapter.bin
- instantid-controlnet.safetensors
- antelopev2
- Disable metadata on saved outputs (#14 by @digitaljohn)
- Add support for AIO_Preprocessor on ComfyUI_Controlnet_Aux
- Allow passing of an already parsed JSON workflow (helpful when forking the repo to push new Replicate models)
Models added:
Loras added:
Models added:
- artificialguybr LoRAs
- epicrealism_naturalSinRC1VAE.safetensors
- motionctrl_svd.ckpt
- sdxl_vae.safetensors
Face restoration and detection models:
- codeformer.pth
- GFPGANv1.3.pth
- GFPGANv1.4.pth
- RestoreFormer.pth
- detection_mobilenet0.25_Final.pth
- detection_Resnet50_Final.pth
- parsing_parsenet.pth
- yolov5l-face.pth
- yolov5n-face.pth
Models added:
Custom nodes added:
Models added:
- photomaker-v1.bin
- albedobaseXL_v13.safetensors
- RealESRGAN_x4plus_anime_6B.pth
- 4x_NMKD-Siax_200k.pth
- 4x-UltraSharp.pth
- Harrlogos_v2.0.safetensors
- starlightXLAnimated_v3.safetensors
Custom nodes added: