Vous avez un workflow ComfyUI ? Obtenez une documentation lisible directement dans votre navigateur.Rendre un workflow lisible.Essayer maintenant ↗
Anima → Krea2 → SeedVR2 | Three-Stage Anime Pipeline (txt2img → repaint → upscale)
Profil d'exécution
Description de la source
Anima → Krea2 → SeedVR2 | Three-Stage Anime Pipeline (txt2img → repaint → upscale)
A single-queue workflow that combines Anima's tag-based anime generation with Krea2 Turbo's natural-language refinement, finished with SeedVR2 upscaling.
What it does
Anima base (txt2img) — Anima base v10 generates the composition from Danbooru-style tag/nature language prompts. Full LoRA support via Power Lora Loader.
Krea2 repaint (img2img) — The base image is passed straight into Krea2 Turbo through the shared qwen_image_vae (no quality loss from format round-trips). 8 steps / CFG 1, denoise 0.3 by default:
0.2–0.35 → light polish, keeps Anima's composition and linework
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.
This workflow is for creating anime images: it generates a base image from Danbooru-style tag prompts with Anima, refines it with a natural-language prompt in Krea2 Turbo, and upscales it with SeedVR2, producing base, repainted, and final images.
Anima base v10 creates the composition, Krea2 Turbo repaints it through img2img, and SeedVR2 7B upscales it.
The workflow saves the base, repaint, and final stages and includes an A/B comparer for the Anima base and Krea2 repaint.
The generated base image feeds directly into Krea2 Turbo through the shared qwen_image_vae.
SeedVR2 applies a lanczos ×0.7 pre-scale, then upscales the image to 2000px.
Required custom nodes include rgthree-comfy for Power Lora Loader, Image Comparer, and Fast Groups Bypasser; ComfyUI-SeedVR2_VideoUpscaler for SeedVR2 DiT/VAE loaders and upscaler; and ConditioningKrea2Rebalance for Krea2 NSFW.
Named model files include anima_baseV10.safetensors, ema_vae_fp16.safetensors, krea2TurboOfficialComfy_krea2TurboFp8.safetensors, and qwen_3_06b_base_anima.safetensors.
Other named model files are qwen_image_vae.safetensors, qwen3vl_4b_bf16.safetensors, and seedvr2_ema_7b_fp16.safetensors.
Named node packs are comfyui-workflow-encrypt, rgthree-comfy, and seedvr2_videoupscaler.
Use tag prompts with Anima and natural-language prompts with Krea2; for the same scene, describe it on both sides for consistency.
Suggestion · non vérifié
Krea2 defaults to 8 steps, CFG 1, and denoise 0.3; denoise 0.2–0.35 is described as light polish, while 0.4–0.55 is described as stronger restyling.
Set the shared seed to fixed to reproduce and iterate; use Fast Groups Bypasser to skip SeedVR2 during previews and reduce VRAM use.
Suggestion · non vérifié
Before running, verify that these named items are available under the exact names: ema_vae_fp16.safetensors, qwen_image_vae.safetensors, krea2TurboOfficialComfy_krea2TurboFp8.safetensors, qwen_3_06b_base_anima.safetensors, seedvr2_ema_7b_fp16.safetensors, and seedvr2_videoupscal
Suggestion · non vérifié
Faut-il modifier cela ?
Connectez-vous pour envoyer une demande de modification.
Sources
1 sourceExtraits de sources
3 extraitsSource context: 96 downloads · Type Workflows · Base model Anima
0.4–0.55 → stronger Krea2 restyling, more detail repainted
SeedVR2 upscale — lanczos ×0.7 pre-scale, then SeedVR2 7B upscales to 2000px.
Features
Separate LoRA loaders and prompts per model — tag prompts for Anima, natural language for Krea2. Describe the same scene on both sides for best consistency.
Shared seed — one Seed Generator feeds both KSamplers, so one roll syncs the whole chain. Set it to fixed to reproduce and iterate.
One-click upscale bypass — Fast Groups Bypasser toggles the SeedVR2 group for fast preview rounds (saves VRAM too).
Built-in A/B comparer — slide between the Anima base and Krea2 repaint.
Intermediate saves at every stage (base / repaint / final), core SaveImage nodes only — workflow embeds in PNG metadata as usual.
Required custom nodes
rgthree-comfy : Power Lora Loader, Image Comparer, Fast Groups Bypasser
ComfyUI-SeedVR2_VideoUpscaler : SeedVR2 DiT/VAE loaders + upscaler
ConditioningKrea2Rebalance : Krea2 NSFW
Estimation des besoins VRAM
Estimation indisponible
12,2 GB sur 2 de 8 fichiers de modèle. Total des fichiers du modèle + 25 % de surcharge de chargement + 2 Go de tampon d'exécution, arrondi à l'unité supérieure.
Exigences
Exigences 10Checkpoint · 3.89 GB · SAFETENSORS · bf16 · full · Unknown
ema_vae_fp16.safetensors
ConflitVAE · Unknown
krea2TurboOfficialComfy_krea2TurboFp8.safetensors
Non résoluCheckpoint · Unknown
qwen_3_06b_base_anima.safetensors
Non résoluText encoder · Unknown
VAE · Unknown
Text encoder · 8.27 GB · SAFETENSORS · bf16 · Unknown
seedvr2_ema_7b_fp16.safetensors
ConflitCheckpoint · Unknown
Sources possibles
seedvr2_videoupscaler
Non résoluUpscaler · Registry
Pack de nœud · Registry
Pack de nœud · Registry
Source context: 95 downloads · Type Workflows · Base model Anima
Source context: 90 downloads · Type Workflows · Base model Anima
