Anima adetailer face workflow — v1.0 — qwenImgv1.0
Detailer workflows that use detailers to fix hands and faces with the Anima and Qwen Image 2.1 models.
Profil d'exécution
- Accès
- Flux de travail externe
- Statut
- Présenté
- Version
- qwen_detailer.json
- Source
- Civitai
- Mis à jour
- 16 oct. 2026
Description de la source
Detailer workflows that use detailers to fix hands and faces with the Anima and Qwen Image 2.1 models.
This is just an example workflow; please feel free to replace the nodes with the versions you are using.
For faces or eyes, Anima ADetailer is sufficient. For hands, feet, or other generation errors, the Qwen Image workflow is recommended.
本工作流利用 Detailer 节点结合 Anima 与 Qwen Image 2.1 模型来修复各种瑕疵。
这只是一个示例工作流,您可以根据自己本地使用的节点版本随意替换。
面部或眼部修复: 用 Anima ADetailer 就足够了。
手部、足部或画面其他崩坏部分: 推荐使用 Qwen Image 修复流。
Note for the Qwen Image workflow:
If you haven't used Qwen Image 2.1 before, please refer to the following ComfyUI documentation .
The detailer fix can be unstable, so you may need to generate multiple times for the best result.
I am using a Chinese prompt to fix hands in this workflow for better results, but you can use the English .
for other prompt please use any LLM to generate with: "search reference and generate a prompt for [ ] with Qwen Image 2.1 in chinese and english.“
Qwen Image 修复流说明:
如果您之前没用过 Qwen Image 2.1,请先参考 ComfyUI 文档 。
Detailer 修复效果有一定随机性,可能需要多次运行才能得到满意结果。
工作流中提供了手部修复提示词,如需其他工作请使用各类LLm生成。
What you need
Hardware and download-size estimates are not available yet.
Exigences
Modèles
4- Non résolu
Object Detector · Hugging Face
qwen_image_2.1_int8_convrot.safetensors
Fichier non vérifiéCheckpoint · Hugging Face · Comfy-Org/Qwen-Image-2.1
qwen_image_2.1_vae_bf16.safetensors
Fichier non vérifiéVAE · Hugging Face · Comfy-Org/Qwen-Image-2.1
- Fichier non vérifié
Text encoder · Hugging Face · Comfy-Org/Qwen-Image-2.1
Packs de nœuds
4- Trouvé
Pack de nœud · Registry
- Trouvé
Pack de nœud · Registry
- Trouvé
Pack de nœud · Registry
Pack de nœud · Registry
Commentaire généré par l’IA
Explication générée par l’IA à partir des détails de la source et de la configuration. Les suggestions sont clairement signalées.
Objectif
Use this ComfyUI workflow to repair an image: it targets faces or eyes with Anima ADetailer, and hands, feet, or other generation errors with Qwen Image 2.1, producing an edited image.
Flux
The graph loads an image, crops areas for inpainting, stitches them back, and includes preview and image-saving nodes.
Entrées et sorties
For hand repair, the workflow includes a Chinese prompt; the description says an English prompt can also be used.
Installation
The author describes this as an example workflow and says you can replace its nodes with versions you use locally.
If you are new to Qwen Image 2.1, the description points to ComfyUI's Qwen Image 2.1 documentation.
Dépendances
The listed node packs are comfyui-image-saver (1.21.0), comfyui-impact-pack (8.28.3), comfyui-impact-subpack (1.3.5), and comfyui-inpaint-cropandstitch (3.0.17).
The listed model files are bbox/hand_yolov8s.pt, qwen_image_2.1_int8_convrot.safetensors, qwen_image_2.1_vae_bf16.safetensors, and qwen3vl_8b_bf16.safetensors.
Configuration
For other repair targets, the description suggests preparing a prompt in Chinese and English.
Suggestion · non vérifié
Précautions
The description says detailer results can vary, so multiple runs may be needed to get a satisfactory result.
Faut-il modifier cela ?
Connectez-vous pour envoyer une demande de modification.
Sources
1 source
Sources
1 sourceExtraits de sources
1 extrait
Extraits de sources
1 extraitSource context: 689 downloads · Type Workflows · Base model Qwen 2.1
