SDXL Cinematic (NSFW) Anime,Portrait Workflow Hires. fix VAE + Ksampler Workflow
5.0
0 reviewsDescription
Workflow Information
Model: Stable Diffusion XL (SDXL) Base 1.0 (or your preferred checkpoint)
Workflow Type: A Hires. fix workflow was used to enhance the image quality.
Key Nodes: This workflow utilizes a CheckpointLoaderSimple for the model, CLIPTextEncode for both positive and negative prompts, a KSampler for the initial generation, and then a LatentUpscale followed by a second KSampler with a lower denoise value for the Hires. fix step. A VAEDecode and SaveImage node complete the process.
Prompting Style: The prompt uses weighted terms to emphasize key elements like photorealism, lighting, and character details, ensuring a highly cinematic and aesthetically pleasing final image.
If your workflows aren’t working anymore, try rolling back to an older version of ComfyUI. The update from about a month ago made mine incompatible
Discussion
(No comments yet)
Loading...
Reviews
No reviews yet
Versions (1)
- latest (5 months ago)
Node Details
Primitive Nodes (14)
Fast Groups Bypasser (rgthree) (1)
Image Comparer (rgthree) (1)
Label (rgthree) (5)
PrimitiveStringMultiline (1)
workflow>DetailerBreast (4)
workflow>DetailerFace (1)
workflow>DetailerHand (1)
Custom Nodes (19)
ComfyUI
- SaveImage (4)
- CLIPTextEncode (2)
- UpscaleModelLoader (1)
- ImageUpscaleWithModel (1)
- ImageScale (1)
- EmptyLatentImage (1)
- LatentUpscale (1)
- VAEDecode (2)
- KSampler (2)
- CheckpointLoaderSimple (1)
- DPCombinatorialGenerator (1)
- StringConcatenate (1)
- SaveText|pysssss (1)
Model Details
Checkpoints (1)
novaAnimeXL_ilV90.safetensors
LoRAs (0)