Skip to content
WorkflowListedExternal
Open source

简单易懂的anima工作流/text2img:Simple Workflow for anima(lora、fix、FaceDetailer) — v0.2 — v0.3

To replace the nodes with those required by Anima-2.9B, we have replaced the LLLite model and added DyPE.

Sourcecivitai.com/models/2763583
VersionRelease 3267957 — Simple Workflow for anima-v0.3.jsonEstimated VRAMNot yet characterized

Runtime profile

Access
External workflow
Status
Listed
Version
Simple Workflow for anima-v0.3.json
Source
Civitai
Updated
Oct 18, 2026

Source description

将节点更换为Anima-2.9B所需要的,更换了LLLite模型,加入了DyPE。

To replace the nodes with those required by Anima-2.9B, we have replaced the LLLite model and added DyPE.

What you need

Model files
0.2 GiB known subtotal

Remaining hardware or download-size estimates are not available yet.

Requirements

Models

8

  • 4x-UltraSharpV2_Lite.safetensors

    File unverified

    Upscaler · Hugging Face · Kim2091/UltraSharpV2

  • animaLLMLayerwisefp16_v1.safetensors

    Not resolved

    Text encoder · Unknown

  • Anzhc Eyes -seg-hd.pt

    Not resolved

    Object Detector · Unknown

  • Anzhc Face seg 640 v4 y11n.pt

    Not resolved

    Object Detector · Unknown

  • controlnetLlliteAnima_ab1CANNY.safetensors

    Found

    ControlNet · 7.48 MB · SAFETENSORS · fp8 · Unknown

  • MatureRitual-anima-0.2.safetensors

    Not resolved

    Checkpoint · Unknown

  • qwen_image_vae.safetensors

    Found

    VAE · 242 MB · SAFETENSORS · Unknown

  • sam_vit_l_0b3195.pth

    Not resolved

    Segmentation · Unknown

Node packs

8

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Purpose

  • This Anima text-to-image workflow turns text prompts into images through a three-step process that includes LoRA, upscaling, and face repair.

Flow

  • Its graph includes text-prompt encoding, sampling, resolution selection, FaceDetailer, UltimateSDUpscale, image preview, and saving.

Inputs and outputs

  • The graph includes PreviewImage and SaveImage stages for previewing and saving generated images.

Dependencies

  • The workflow lists these node packs: comfyui_controlnet_aux, comfyui_essentials, comfyui_ultimatesdupscale, ComfyUI-DyPE, comfyui-easy-use, comfyui-impact-pack, comfyui-impact-subpack, and rgthree-comfy.

  • Listed model files include 4x-UltraSharpV2_Lite.safetensors, animaLLMLayerwisefp16_v1.safetensors, Anzhc Eyes -seg-hd.pt, and Anzhc Face seg 640 v4 y11n.pt.

  • Other listed model files are controlnetLlliteAnima_ab1CANNY.safetensors, MatureRitual-anima-0.2.safetensors, qwen_image_vae.safetensors, and sam_vit_l_0b3195.pth.

Configuration

  • You can adjust the step count and upscaling factor to suit your needs.

    Suggestion · not verified

  • The default model is for testing only; the creator says you may use any model you prefer.

Cautions

  • The creator reports 450–570 seconds on an RTX 3060 with the default steps and upscaling factor; this is a source-reported figure, not a tested runtime here.

  • Before installing, verify the exact packages for ComfyUI-DyPE and comfyui_controlnet_aux; the listing provides no package links for them.

    Suggestion · not verified

Does this need to be edited?

Sign in to send an edit request.

Sources

1 source
CivitaiPrimaryVersion 3267957
civitai.com/models/2763583

Verified Oct 18

Source excerpts

1 excerpt
NoteOct 18civitai.com/models/2763583

Source context: 718 downloads · Type Workflows · Base model Anima