Wan2.2 - Ultimate SD Image Upscaler - Batch loader - Florence - Master UI
Perfil de execução
Descrição da fonte
Wan2.2 - Ultimate SD Image Upscaler - Batch loader - Florence - Master UI
Upscale Model - https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth
Clip -
Comentário gerado por IA
Explicação gerada por IA com base nos detalhes da fonte e da configuração. As sugestões são claramente identificadas.
This workflow is for image upscaling. It can load single images or image batches, process prompt text with Florence, and save image outputs; its graph also includes WanImageToVideo.
The listed steps cover image loading, batch loading, prompt replacement, text concatenation, Florence runs, scaling, model-based upscaling, image comparison, and saving.
The graph also includes CLIP loading, text encoding, resolution controls, and WanImageToVideo.
LoadImage and Load Image Batch identify single-image and batch-image inputs.
CLIPTextEncode, Text Concatenate, and Florence2Run are listed for text processing; SaveImage is listed for image output, and Image Comparer (rgthree) is listed for image comparison.
The workflow is listed on Civitai as a Workflow and is published by Sinusplusminus.
Listed model files are lightx2v_T2V_14B_cfg_step_distill_v2_lora_rank64_bf16_.safetensors, MiaoshouAI/Florence-2-base-PromptGen-v2.0, RealESRGAN_x2plus.pth, umt5_xxl_fp8_e4m3fn_scaled.safetensors, wan_2.1_vae.safetensors, and Wan2_2-T2V-A14B-LOW_fp8_e4m3fn_scaled_KJ.safetensors.
Listed node packs are comfyui_tinyterranodes, comfyui_ultimatesdupscale, comfyui-easy-use, comfyui-florence2, comfyui-workflow-encrypt, crt-nodes, and was-ns.
The listed base model is Wan Video 2.2 I2V-A14B.
Listed node-pack pins are comfyui_tinyterranodes 2.0.9, comfyui_ultimatesdupscale 1.5.1, comfyui-easy-use 5c8cdb58c798a404e50636c6f3e179b084a23fab, and comfyui-florence2 1.0.3.
The remaining listed pins are comfyui-workflow-encrypt 1.0.0, crt-nodes 1.9.6, and was-ns 3.0.1.
Before setup, verify the source and exact file for wan_2.1_vae.safetensors, lightx2v_T2V_14B_cfg_step_distill_v2_lora_rank64_bf16_.safetensors, Wan2_2-T2V-A14B-LOW_fp8_e4m3fn_scaled_KJ.safetensors, and umt5_xxl_fp8_e4m3fn_scaled.safetensors; no source URL is listed for these four
Sugestão · não verificado
Before setup, confirm the exact comfyui-florence2 pack and RealESRGAN_x2plus.pth file, and verify where to obtain was-ns.
Sugestão · não verificado
Isso precisa ser editado?
Faça login para enviar uma solicitação de edição.
Fontes
1 fonteTrechos de fonte
1 trechoSource context: 204 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B
Estimativa de requisito VRAM
Estimativa indisponível
14,0 GB distribuídos em 1 de 7 arquivos de modelo. Total de arquivos do modelo + 25% de overhead de carregamento + 2 GB de buffer de execução, arredondado para cima.
Requisitos
Requisitos 13Upscaler · Registry · ComfyUI_UltimateSDUpscale
lightx2v_T2V_14B_cfg_step_distill_v2_lora_rank64_bf16_.safetensors
Não resolvidoLoRA · Unknown
MiaoshouAI/Florence-2-base-PromptGen-v2.0
Arquivo não verificadoVision model · Hugging Face · MiaoshouAI/Florence-2-base-PromptGen-v2.0 · Florence 2 Base PromptGen v2.0
RealESRGAN_x2plus.pth
ConflitoUpscaler · Unknown
umt5_xxl_fp8_e4m3fn_scaled.safetensors
Arquivo não verificadoText encoder · SAFETENSORS · Unknown
VAE · Unknown
Checkpoint · 14.0 GB · SAFETENSORS · fp8 · pruned · Unknown
Pacote de nós · Registry
Pacote de nós · Registry
comfyui-florence2
Pacote de nós · Registry
Pacote de nós · Registry
Pacote de nós · Registry
was-ns
Não resolvidoPacote de nós · Registry