TokiToki Internal
Atlas + S3
All runs
CompletedComfyUI

b5ae7d53-d8c1-4746-b627-b757e32547fa

Queued
Jul 27, 2026, 03:10 PM
Completed
Jul 27, 2026, 03:10 PM
Execution time
21.4s
Outputs
1 · 5.6 MB
Inputs
2
Cached nodes

Outputs

1

Click an image to view it full size.

character_ref_00552_.png

node 3 · 5.6 MB

Text prompts

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

Tight insert: @Aria's hand slapping flat against a wet steel lamppost and gripping to shove off it mid-run, knuckles tensed, water flinging off the back of her hand and off the cold metal. Set: the dark street of @RainStreet/Road, only the wet lamppost and black bokeh behind. Composition: tight, the hand and a section of the rain-slick post filling the frame, entering from the running direction, motion-implied. Camera: shot on ARRI Alexa 35, Signature Prime 65mm at T2.8, shallow focus on the hand and metal. Lighting: cold streetlight edge glancing off the wet post and skin, hard specular highlights, deep shadow. Color: ARRI Reveal, desaturated cool, deep blacks. Texture: real skin on the hand, water droplets flicking, condensation and rain on the steel, believable metal, no smoothing. Photorealistic cinematic still, prestige crime-drama grit, 9:16.

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

2

Source images and video this run consumed.

character_ref_00545_.png

node 4 · 6.2 MB

character_ref_00538_.png

node 5 · 6.6 MB