### Start ComfyUI Web Server Inside Cog Container Source: https://github.com/fofr/cog-comfyui-supir/blob/main/README.md Once inside the Cog container, this command navigates to the ComfyUI directory and starts the main Python server, binding it to all network interfaces. This makes the ComfyUI Web UI accessible. ```shell cd ComfyUI/ python main.py --listen 0.0.0.0 ``` -------------------------------- ### Install Custom Nodes for Cog ComfyUI Source: https://github.com/fofr/cog-comfyui-supir/blob/main/README.md This script installs all custom nodes required for the Cog ComfyUI project. It ensures that your local development environment has all necessary components. ```shell ./scripts/install_custom_nodes.py ``` -------------------------------- ### Start Cog Container and Expose Port for ComfyUI Web UI Source: https://github.com/fofr/cog-comfyui-supir/blob/main/README.md This command starts the Cog container and exposes port 8188, allowing access to the ComfyUI Web UI from outside the container. This is necessary for running the Web UI on a GPU machine. ```shell sudo cog run -p 8188 bash ``` -------------------------------- ### ComfyUI SUPIR CLIP Models List Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md Details the available CLIP (Contrastive Language-Image Pre-training) models. These models are used for understanding and processing text-image relationships, often for prompt encoding or image-text similarity tasks, crucial for guiding image generation. ```APIDOC CLIP: - models--QuanSun--EVA-CLIP ``` -------------------------------- ### InstantID Model File Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section details the InstantID IP-Adapter model file. This model is likely used for identity preservation in image generation tasks and is provided in a .bin format. ```APIDOC instantid-ip-adapter.bin ``` -------------------------------- ### Clone Cog ComfyUI Repository with Submodules Source: https://github.com/fofr/cog-comfyui-supir/blob/main/README.md This command clones the `cog-comfyui` repository from GitHub, including all its submodules. This is the first step for local development. ```shell git clone --recurse-submodules https://github.com/fofr/cog-comfyui.git ``` -------------------------------- ### GroundingDino Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section details GroundingDino models, which are typically used for open-set object detection by grounding text prompts. These models enable flexible and powerful object recognition capabilities. ```APIDOC groundingdino_swinb_cogcoor.pth groundingdino_swint_ogc.pth ``` -------------------------------- ### PhotoMaker Model File Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This entry specifies the PhotoMaker model file, which is used for generating personalized images. The model is provided in a .bin format. ```APIDOC photomaker-v1.bin ``` -------------------------------- ### AnimateDiff Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section lists various AnimateDiff models for generating animated sequences. It includes different versions, motion modules, and lightning models compatible with ComfyUI and Diffusers, offering diverse animation capabilities. ```APIDOC mm_sd_v14.ckpt mm_sd_v15.ckpt mm_sd_v15_v2.ckpt mm_sdxl_v10_beta.ckpt v3_sd15_adapter.ckpt v3_sd15_mm.ckpt v3_sd15_sparsectrl_rgb.ckpt v3_sd15_sparsectrl_scribble.ckpt temporaldiff-v1-animatediff.ckpt mm-Stabilized_high.pth mm-Stabilized_mid.pth lt_long_mm_32_frames.ckpt lt_long_mm_16_64_frames.ckpt lt_long_mm_16_64_frames_v1.1.ckpt sd15_t2v_beta.ckpt hsxl_temporal_layers.f16.safetensors hsxl_temporal_layers.safetensors 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 animatediffLCMMotion_v10.ckpt AnimateLCM_sd15_t2v.ckpt ``` -------------------------------- ### PuLID Model File Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section details the PuLID IP-Adapter model for SDXL. This model is likely used for personalized image generation with strong identity preservation capabilities. ```APIDOC ip-adapter_pulid_sdxl_fp16.safetensors ``` -------------------------------- ### UNet Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section lists various UNet model files, including different step versions for lightning models. These models are typically used in diffusion models for image generation and are available in both .pth and .safetensors formats. ```APIDOC sdxl_lightning_1step_unet_x0.pth sdxl_lightning_1step_unet_x0.safetensors sdxl_lightning_2step_unet.pth sdxl_lightning_2step_unet.safetensors sdxl_lightning_4step_unet.pth sdxl_lightning_4step_unet.safetensors sdxl_lightning_8step_unet.pth sdxl_lightning_8step_unet.safetensors ``` -------------------------------- ### ComfyUI SUPIR Checkpoints List Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md Lists the various Stable Diffusion and other base model checkpoints available for image generation and manipulation within the ComfyUI SUPIR environment. These models are typically used as the primary generative backbone for creating images from text prompts. ```APIDOC Checkpoints: - 512-inpainting-ema.safetensors - absolutereality_v181.safetensors - albedobaseXL_v13.safetensors - albedobaseXL_v21.safetensors - anything-v3-fp16-pruned.safetensors - CinematicRedmond.safetensors - copaxCuteXLSDXL10_v4.safetensors - copaxTimelessxlSDXL1_v8.safetensors - Deliberate_v2.safetensors - dreamlabsoil_V2_v2.safetensors - DreamShaper_6.2_BakedVae_pruned.safetensors - DreamShaper_6.31_BakedVae.safetensors - DreamShaper_6.31_BakedVae_pruned.safetensors - DreamShaper_6.31_INPAINTING.inpainting.safetensors - DreamShaper_6_BakedVae.safetensors - dreamshaper_8.safetensors - dreamshaper_8LCM.safetensors - dreamshaperXL_alpha2Xl10.safetensors - dreamshaperXL_lightningDPMSDE.safetensors - dynavision_v20Bakedvae.safetensors - epicrealism_naturalSinRC1VAE.safetensors - epicrealismXL_v10.safetensors - Hyper-SDXL-1step-Unet-Comfyui.fp16.safetensors - Hyper-SDXL-1step-Unet.safetensors - imp_v10.safetensors - jibMixRealisticXL_v10Lightning46Step.safetensors - Juggernaut-XL_v9_RunDiffusionPhoto_v2.safetensors - juggernaut_reborn.safetensors - Juggernaut_RunDiffusionPhoto2_Lightning_4Steps.safetensors - juggernautXL_juggernautX.safetensors - juggernautXL_v8Rundiffusion.safetensors - LCM_Dreamshaper_v7_4k.safetensors - magicmixReverie_v10.safetensors - majicmixRealistic_v7.safetensors - motionctrl.pth - motionctrl_svd.ckpt - photonLCM_v10.safetensors - pixlAnimeCartoonComic_v10.safetensors - playground-v2.5-1024px-aesthetic.fp16.safetensors - proteus_v02.safetensors - ProteusV0.4-Lighting.safetensors - ProteusV0.4.safetensors - rcnzCartoon3d_v20.safetensors - Realistic_Vision_V5.1-inpainting.ckpt - Realistic_Vision_V5.1-inpainting.safetensors - Realistic_Vision_V5.1.ckpt - Realistic_Vision_V5.1.safetensors - Realistic_Vision_V5.1_fp16-no-ema-inpainting.ckpt - Realistic_Vision_V5.1_fp16-no-ema-inpainting.safetensors - Realistic_Vision_V5.1_fp16-no-ema.ckpt - Realistic_Vision_V5.1_fp16-no-ema.safetensors - Realistic_Vision_V6.0_NV_B1.safetensors - Realistic_Vision_V6.0_NV_B1_fp16.safetensors - Realistic_Vision_V6.0_NV_B1_inpainting.safetensors - Realistic_Vision_V6.0_NV_B1_inpainting_fp16.safetensors - realisticLCMBYStable_v10.safetensors - RealVisXL_V2.0.safetensors - RealVisXL_V3.0.safetensors - RealVisXL_V3.0_Turbo.safetensors - RealVisXL_V4.0.safetensors - RealVisXL_V4.0_Lightning.safetensors - rundiffusionXL_beta.safetensors - sd_xl_base_1.0.safetensors - sd_xl_base_1.0_0.9vae.safetensors - sd_xl_refiner_1.0.safetensors - sd_xl_refiner_1.0_0.9vae.safetensors - sd_xl_turbo_1.0.safetensors - sd_xl_turbo_1.0_fp16.safetensors - sdxl_lightning_1step_x0.safetensors - sdxl_lightning_2step.safetensors - sdxl_lightning_4step.safetensors - sdxl_lightning_8step.safetensors - sdxlUnstableDiffusers_nihilmania.safetensors - sdxlUnstableDiffusers_v11Rundiffusion.safetensors - segmind-vega.safetensors - SSD-1B.safetensors - starlightXLAnimated_v3.safetensors - svd.safetensors - svd_xt.safetensors - toonyou_beta6.safetensors - turbovisionxlSuperFastXLBasedOnNew_tvxlV32Bakedvae.safetensors - v1-5-pruned-emaonly.ckpt - v2-1_512-ema-pruned.safetensors - v2-1_768-ema-pruned.ckpt - v2-1_768-ema-pruned.safetensors - v2-1_768-nonema-pruned.ckpt - v2-1_768-nonema-pruned.safetensors - wd-illusion-fp16.safetensors - x4-upscaler-ema.safetensors ``` -------------------------------- ### InsightFace Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section lists various InsightFace models, including those for face recognition (antelopev2, buffalo_l) and face swapping (inswapper). These models are available in .onnx format and are crucial for facial analysis tasks. ```APIDOC antelopev2 buffalo_l inswapper_128.onnx inswapper_128_fp16.onnx models/antelopev2 models/buffalo_l ``` -------------------------------- ### Segment Anything Model (SAM) Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section lists various Segment Anything Models (SAM), including mobile and HQ versions. These models are designed for robust object segmentation across diverse images. ```APIDOC mobile_sam.pt sam_hq_vit_b.pth sam_hq_vit_h.pth sam_hq_vit_l.pth sam_vit_b_01ec64.pth sam_vit_h_4b8939.pth sam_vit_l_0b3195.pth ``` -------------------------------- ### Face Detection Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section details various models for face detection and parsing, including MobileNet, ResNet, BiSeNet, ParseNet, and YOLOv5-face. These models are essential for identifying and analyzing facial features in images. ```APIDOC detection_mobilenet0.25_Final.pth detection_Resnet50_Final.pth parsing_bisenet.pth parsing_parsenet.pth yolov5l-face.pth yolov5n-face.pth ``` -------------------------------- ### ComfyUI SUPIR Upscale Models List Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md Provides a list of models specifically designed for upscaling images, enhancing resolution and detail. These models are used to improve the quality of generated or existing images, often after initial generation by a checkpoint model. ```APIDOC Upscale models: - 4x-AnimeSharp.pth - 4x-UltraMix_Balanced.pth - 4x-UltraMix_Smooth.pth - 4x-UltraSharp.pth - 4x_foolhardy_Remacri.pth - 4x_NMKD-Siax_200k.pth - 8x_NMKD-Superscale_150000_G.pth - ESRGAN_4x.pth - RealESRGAN_x2.pth - RealESRGAN_x4.pth - RealESRGAN_x4plus.pth - RealESRGAN_x4plus_anime_6B.pth - RealESRGAN_x8.pth ``` -------------------------------- ### REMBG (Remove Background) Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section lists various models for background removal, including ISNet and U2Net variants. These models are often used for image matting and producing clean cutouts. ```APIDOC isnet-anime.onnx isnet-general-use.onnx silueta.onnx u2net.onnx u2net_cloth_seg.onnx u2net_human_seg.onnx u2netp.onnx vit_b-decoder-quant.onnx vit_b-encoder-quant.onnx ``` -------------------------------- ### LayerDiffusion Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section lists LayerDiffusion models for various tasks like background/foreground separation and transparent attention. These models are available for both SD1.5 and SDXL architectures, enabling advanced image manipulation. ```APIDOC 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 ``` -------------------------------- ### Ultralytics YOLO Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section provides Ultralytics YOLO model files for various detection and segmentation tasks. It includes models for face, hand, hair, and person detection or segmentation, optimized for different purposes. ```APIDOC bbox/face_yolov8m.pt bbox/hand_yolov8s.pt segm/hair_yolov8n-seg_60.pt segm/person_yolov8m-seg.pt ``` -------------------------------- ### CLIP Segmentation Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section specifies the CLIP-based segmentation model from CIDAS. This model leverages CLIP's understanding of images and text for robust segmentation tasks. ```APIDOC models--CIDAS--clipseg-rd64-refined ``` -------------------------------- ### AnimateDiff LORA Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section details AnimateDiff LORAs (Low-Rank Adaptation) for specific camera movements and animations. These LORAs enable fine-grained control over motions like panning, rolling, tilting, and zooming in generated animations. ```APIDOC v2_lora_PanLeft.ckpt v2_lora_PanRight.ckpt v2_lora_RollingAnticlockwise.ckpt v2_lora_RollingClockwise.ckpt v2_lora_TiltDown.ckpt v2_lora_TiltUp.ckpt v2_lora_ZoomIn.ckpt v2_lora_ZoomOut.ckpt ``` -------------------------------- ### Face Restoration Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section lists various models dedicated to face restoration, such as CodeFormer, GFPGAN, and RestoreFormer. These models are designed to enhance and repair facial images, improving their quality and detail. ```APIDOC codeformer.pth GFPGANv1.3.pth GFPGANv1.4.pth RestoreFormer.pth ``` -------------------------------- ### MMDets Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section specifies MMDetection models, including a YOLOv3 model specifically for anime face detection. These models are part of the MMDetection framework for object detection. ```APIDOC bbox/mmdet_anime-face_yolov3.pth ``` -------------------------------- ### ComfyUI SUPIR CLIP Vision Models List Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md Lists the CLIP Vision models, which are components of CLIP specifically focused on image encoding. They are used to extract visual features from images for various AI tasks, such as image-to-text generation or image analysis. ```APIDOC CLIP Vision: - CLIP-ViT-bigG-14-laion2B-39B-b160k.safetensors - CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors - clip-vit-large-patch14.bin - clip_vision_g.safetensors - IPAdapter_image_encoder_sd15.safetensors - model.15.safetensors - model.sdxl.safetensors ``` -------------------------------- ### Frame Interpolation Model Files Source: https://github.com/fofr/cog-comfyui-supir/blob/main/supported_weights.md This section lists various models for frame interpolation, including RIFE, FILM, AMT, and CAIN. These models are used to generate intermediate frames between existing ones, creating smoother video sequences. ```APIDOC rife40.pth rife41.pth rife42.pth rife43.pth rife44.pth rife45.pth rife46.pth rife47.pth rife48.pth rife49.pth sudo_rife4_269.662_testV1_scale1.pth film_net_fp32.pt amt-s.pth amt-g.pth amt-l.pth gopro_amt-s.pth pretrained_cain.pth ``` -------------------------------- ### Reference Files from Extracted Zip/Tar Archive in ComfyUI Workflow Source: https://github.com/fofr/cog-comfyui-supir/blob/main/README.md When a zip or tar file is uploaded, its contents are extracted to the `input` directory. This snippet illustrates how to reference individual files and directories from within the extracted archive in your ComfyUI workflow's API JSON. ```json "image": "my_img.png", ... "directory": "references", ``` -------------------------------- ### Reference Single Uploaded Input File in ComfyUI Workflow Source: https://github.com/fofr/cog-comfyui-supir/blob/main/README.md When uploading a single input file to ComfyUI, it is saved as `input.[extension]` in the `input` directory. This snippet demonstrates how to modify your workflow's API JSON to reference this uploaded file directly. ```diff - "image": "/your-path-to/image.jpg", + "image": "image.jpg", ``` -------------------------------- ### Modify ComfyUI API JSON to Use URL for Image Input Source: https://github.com/fofr/cog-comfyui-supir/blob/main/README.md This snippet shows how to update your ComfyUI API JSON to reference an image from a URL instead of a local path. This is useful when your input images are hosted externally. ```diff - "image": "/your-path-to/image.jpg", + "image": "https://example.com/image.jpg", ``` === COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.