Loading
Tem um workflow do ComfyUI? Gere uma documentação legível direto no navegador.Torne um workflow legível.Experimentar agora ↗
AP3-to-Illustrious Pixel Relay Upscale Workflow
Perfil de execução
Descrição da fonte
AP3-to-Illustrious Pixel Relay Upscale Workflow
A toy two-stage ComfyUI workflow made for myself that uses Anima Preview 3 to create the base image, then uses Illustrious SDXL to upscale and refine it.
The main idea is to keep AP3 focused on what it does well: caption following, scene setup, and multi-character interaction . At the same time, it avoids relying on AP3 for high-resolution diffusion refinement, since its official recommendation is to stay around the ~1MP range and image quality can degrade at higher resolutions.
In this workflow, AP3 generates the initial base image . The image is then passed in pixel space to the second stage, where the upscale model handles the actual image enlargement , and Illustrious SDXL redraws/refines the enlarged image through UltimateSDUpscale. This avoids pushing AP3 beyond its comfortable resolution range while still taking advantage of Illustrious’s stronger high-resolution redraw, background rendering, and detail refinement .
Comentário gerado por IA
Explicação gerada por IA com base nos detalhes da fonte e da configuração. As sugestões são claramente identificadas.
Use this two-stage ComfyUI workflow with a scene or multi-character caption to create an Anima Preview 3 base image, then enlarge and refine it with Illustrious SDXL; it produces the base image and a refined, enlarged image.
Anima Preview 3 creates the initial base image and supports caption following, scene setup, and multi-character interaction.
The base image moves in pixel space to an upscale stage, where the upscale model enlarges it and Illustrious SDXL redraws and refines it through UltimateSDUpscale.
The second stage supports high-resolution redraw, background rendering, and detail refinement.
Input: a caption for scene setup or multi-character interaction.
Outputs: the initial base image and an enlarged image redrawn and refined by Illustrious SDXL.
Download version v1.0 from Civitai at https://civitai.com/models/2607495?modelVersionId=2927851; it is published by Auleric.
Model files named for this workflow include 4x-UltraSharp.pth, anima-preview3-base.safetensors, qwen_3_06b_base_anima.safetensors, qwen_image_vae.safetensors, and waiIllustriousSDXL_v170.safetensors.
Node packs named for this workflow include comfyui_ultimatesdupscale, comfyui-image-saver, comfyui-mxtoolkit, and rgthree-comfy.
Keep Anima Preview 3 around the ~1MP range for base generation, then use the second stage for enlargement and high-resolution refinement.
Sugestão · não verificado
Before use, verify that anima-preview3-base.safetensors, qwen_image_vae.safetensors, and qwen_3_06b_base_anima.safetensors are available locally and are the versions you intend to use.
Sugestão · não verificado
No license is listed for this workflow.
Isso precisa ser editado?
Faça login para enviar uma solicitação de edição.
Fontes
1 fonteTrechos de fonte
3 trechosSource context: 129 downloads · Type Workflows · Base model Anima
Cover generated by GPT image.
Remaining hardware or download-size estimates are not available yet.
Requisitos
Upscaler · 63.9 MB · PTH · Hugging Face · Kim2091/UltraSharp
anima-preview3-base.safetensors
EncontradoCheckpoint · 3.89 GB · SAFETENSORS · bf16 · full · Civitai · 2458426
Source context: 97 downloads · Type Workflows · Base model Anima
Source context: 90 downloads · Type Workflows · Base model Anima
qwen_3_06b_base_anima.safetensors
Não resolvidoText encoder · Unknown
VAE · 242 MB · Civitai · 1842123
waiIllustriousSDXL_v170.safetensors
Não resolvidoCheckpoint · Civitai · 827184
comfyui_ultimatesdupscale
EncontradoPacote de nós · Registry
comfyui-image-saver
EncontradoPacote de nós · Registry
comfyui-mxtoolkit
EncontradoPacote de nós · Registry
rgthree-comfy
EncontradoPacote de nós · Registry
