TokiToki Internal
Atlas + S3
All runs
CompletedComfyUI

7263f784-48d1-4592-a61b-31a08b336e38

Queued
Jul 27, 2026, 12:34 PM
Completed
Jul 27, 2026, 12:34 PM
Execution time
7.6s
Outputs
1 · 1.7 MB
Inputs
0
Cached nodes

Outputs

1

Click an image to view it full size.

character_ref_00518_.png

node 3 · 1.7 MB

Text prompts

2
Promptnode 1 · PrimitiveStringMultiline · valueCharacter Description (your input)

Cardboard box, A standard, medium-sized, natural light brown cardboard moving box. It is half-filled with a jumble of personal items, suggesting a recent packing or unpacking., The box is constructed from standard corrugated cardboard with a matte finish. It shows visible signs of moderate wear, including faint scuff marks, creased corners, and residual adhesive patches from previously removed packing tape. One of the top flaps is partially ripped. Inside, visible through the open top, are various personal effects: a hardcover book with a faded blue cover, a small, tarnished silver pocket watch peeking from under a folded, plaid scarf, and a stack of old, yellowed letters tied with a thin, brittle string. A small, framed photograph, turned face down, partially covers some of the items, preventing clear identification of the image.

Systemnode 2 · GeminiImageDirect · system_promptCharacter Image Generator (Nano Banana)

Generate a single, hyper-realistic character reference image (strictly no collages, text, watermarks, or borders). The image must feature a cinematic, three-quarter portrait of the described character, maintaining a sharp, tack-focused emphasis on the face with natural skin textures, pore details, and realistic micro-expressions. The lighting should be cinematic and directional (e.g., subtle split or Rembrandt lighting) to give depth, blending seamlessly with the environment specified in the character description. The character must have a natural, living presence, looking slightly off-camera or directly at the lens with a subtle, context-aware expression. Ensure the full head, neck, and upper shoulders are completely unobstructed. This image serves as a high-fidelity identity blueprint for temporal-consistent video generation.

Input media

0

This run had no input media — it generated from prompts alone.