工作流分為一次性完成與兩段式。一次性完成的工作流對顯示卡記憶體的要求最低為12GB。

使用巨大頭部的動漫人物效果會較差。

人物臉部太小時或照片尺寸太小時，FaceDetailer的功能可能會失效。

Workflows are divided into one-time completion and two-stage completion.

One-time completion workflows require a minimum of 12GB of graphics card memory.

Anime characters with huge heads tend to look less appealing.

FaceDetailer may malfunction if the face is too small or the photo size is too small.

範例中使用的LORA:

flux2-klein-9b-real-slider-v2

klein_9B_Turbo_r128

下載Lora Manager.

Download Lora Manager.

下載z-image turbo和flux2 klein 9b base模型。

z_image_turbo_fp8_e4m3fn.safetensors

flux-2-klein-base-9b-fp8.safetensors

先安裝缺失的節點.

下載face_yolov9c.pt

https://huggingface.co/Bingsu/adetailer/blob/main/face_yolov9c.pt

將face_yolov9c.pt放入以下路徑:ComfyUI\models\ultralytics\bbox

--

下載sam_vit_b_01ec64.pth

將sam_vit_b_01ec64.pth放入以下路徑:ComfyUI\models\sams

--

如果面部修復功能沒有生效時，需提高"guide size"的數字.

"FaceDetailer"安裝有錯誤時，請參考"ComfyUI-Impact-Pack"中的教學.

LORA used in the example:

flux2-klein-9b-real-slider-v2

klein_9B_Turbo_r128

Download the z-image turbo and flux2 klein 9b base models.

z_image_turbo_fp8_e4m3fn.safetensors

flux-2-klein-base-9b-fp8.safetensors

Install the missing nodes first.

Download face_yolov9c.pt

https://huggingface.co/Bingsu/adetailer/blob/main/face_yolov9c.pt

Place face_yolov9c.pt in the following path:ComfyUI\models\ultralytics\bbox

--

Download sam_vit_b_01ec64.pth

Place sam_vit_b_01ec64.pth in the following path: ComfyUI\models\sams

--

If the facial repair function is ineffective, the "guide size" number needs to be increased.

--

If you encounter an error during the installation of "FaceDetailer", please refer to the tutorial in "ComfyUI-Impact-Pack".

--

我的其他檢查站和Lora。

My other checkpoints and Lora.

--

客製化lora:我的Telegram ID:@iamddtla0620

我的微信ID:iamddtla

我的line ID:iamddtla

時間寶貴，請直接說明來意，例如:訂製lora.

Customized lora: My Telegram ID: @iamddtla0620

My WeChat ID: iamddtla

My Line ID: iamddtla

Time is precious, please state your intention directly, for example: customized LORA.