Playground 2.5 + SD 1.5
Laufzeitprofil
Quellenbeschreibung
Here's a ComfyUI workflow for the Playground AI - Playground 2.5 model with Face Detailer.
Initially, I considered using the Playground model for the Face Detailer as well, but after extensive testing, I decided to opt for an SD_1.5 model as it yielded the best results for faces, especially in terms of skin appearance.
The Face Detailer can generate up to 5 different faces, each with its own prompt.
Enjoy the workflow :)
And please leave a review if you like the workflow!
Geschätzter VRAM Bedarf
KI-generierter Kommentar
KI-generierte Erklärung auf Grundlage von Quellen- und Konfigurationsdetails. Vorschläge sind klar gekennzeichnet.
Use this ComfyUI workflow to generate Playground 2.5 images with Face Detailer: enter up to five separate face prompts and receive generated faces.
The workflow combines the Playground 2.5 model with an SD_1.5 model for Face Detailer.
Give each face its own prompt; Face Detailer can generate up to five different faces.
Download version v1.0 from the Civitai page and open it in ComfyUI.
Vorschlag · nicht geprüft
Named node components include MaskPreview+, CLIPSetLastLayer, ModelSamplingContinuousEDM, SaveImage, ImpactSimpleDetectorSEGS, CLIPTextEncode, EmptyLatentImage, CheckpointLoaderSimple, UltralyticsDetectorProvider, DetailerForEach, ImpactSEGSOrderedFilter, SAMLoader, KSampler (Eff
Listed model files include playground-v2.5-1024px-aesthetic.fp16.safetensors, epicrealism_naturalSinRC1VAE.safetensors, epicrealism_pureEvolutionV5-inpainting.safetensors, juggernautXL_v7Rundiffusion.safetensors, bbox/face_yolov8n_v2.pt, and sam_vit_b_01ec64.pth.
The workflow requires Adetailer, Juggernaut XL, and epiCRealism (pure Evolution-inpainting).
Use the SD_1.5 model rather than the Playground model for Face Detailer; the description says this choice was made for face results, especially skin appearance.
Before setup, verify the source and filename for MaskPreview+, playground-v2.5-1024px-aesthetic.fp16.safetensors, and epicrealism_naturalSinRC1VAE.safetensors; no provider identifier or checksum is listed for these items.
Vorschlag · nicht geprüft
Review the Civitai page for setup details before use; the listing has no separate description.
Vorschlag · nicht geprüft
Muss das bearbeitet werden?
Melden Sie sich an, um eine Änderungsanfrage zu senden.
Quellen
1 QuelleQuellenauszüge
1 AuszugSource context: 2034 downloads · Type Workflows · Base model Playground v2
Schätzung nicht verfügbar
364 MB über 2 von 6 Modell-Dateien. Gesamtmodell-Dateien + 25% Ladeaufwand + 2 GB Ausführungs-Puffer, aufgerundet.
Anforderungen
23 AnforderungenObject Detector · 5.95 MB · Hugging Face · Bingsu/adetailer
epicrealism_naturalSinRC1VAE.safetensors
Nicht aufgelöstVAE · Unknown
epicrealism_pureEvolutionV5-inpainting.safetensors
Nicht aufgelöstCheckpoint · Civitai · 25694
juggernautXL_v7Rundiffusion.safetensors
Nicht aufgelöstCheckpoint · Civitai · 133005
playground-v2.5-1024px-aesthetic.fp16.safetensors
Nicht aufgelöstCheckpoint · Unknown
Segmentation · 358 MB · Hugging Face · comfyuistudio/sam_vit_b_01ec64.pth
CheckpointLoaderSimple
MöglichKnotenpaket · Unknown
CLIPSetLastLayer
MöglichKnotenpaket · Unknown
CLIPTextEncode
MöglichKnotenpaket · Unknown
Knotenpaket · Unknown
EmptyLatentImage
MöglichKnotenpaket · Unknown
ImageSharpen
MöglichKnotenpaket · Unknown
Knotenpaket · Unknown
Knotenpaket · Unknown
KSampler (Efficient)
MöglichKnotenpaket · Unknown
Knotenpaket · Unknown
ModelSamplingContinuousEDM
MöglichKnotenpaket · Unknown
PreviewImage
MöglichKnotenpaket · Unknown
Knotenpaket · Unknown
SaveImage
MöglichKnotenpaket · Unknown
Knotenpaket · Unknown
Knotenpaket · Unknown
Knotenpaket · Unknown