TokiToki Internal
Atlas + S3
All runs
CompletedComfyUI

6daeefd7-9a08-4631-9a9d-db9e406ecd7e

Queued
Jul 27, 2026, 02:09 PM
Completed
Jul 27, 2026, 02:09 PM
Execution time
12.4s
Outputs
1 · 6.6 MB
Inputs
1
Cached nodes
1

Outputs

1

Click an image to view it full size.

character_ref_00538_.png

node 3 · 6.6 MB

Text prompts

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

Establishing location plate, empty of people. Night, around 8 PM, heavy rain: a hard steady downpour, visible rain streaks in the air, rain bouncing off the pavement, water sheeting down every surface. A wide, completely empty two-lane city road in Lower Manhattan. Wet black asphalt with deep puddles and long stretched reflections of the streetlights, worn painted lane markings, a manhole leaking a column of steam. No vehicles anywhere, no parked cars blocking the frame, the road bare and open (her running lane). On the LEFT, wedged between two tall brick buildings, the dark narrow mouth of a pedestrian alley, far too narrow for any vehicle, a vertical slot of deeper shadow with a single faint light glowing somewhere inside it. Near the alley mouth on the sidewalk stands an NYC subway stair entrance with a glowing green globe lamp and metal railings. Buildings: weathered brownstone and old warehouse facades, rusted fire escapes, a shuttered bodega with a faint buzzing neon sign further down the block. Sky: black and overcast, full night, no daylight, no blue-hour glow. Atmosphere: cold and wet, near-black desaturated palette of slate and grey, broken only by warm pools under the sodium streetlights, the green subway lamp and the distant bodega neon. Volumetric rain, drifting mist, everything soaked and reflective. Lighting entirely motivated by these practical sources, no artificial fill. Photorealistic cinematic location still, prestige crime-drama grit, shot on ARRI Alexa 35 with a wide lens, deep clean blacks, ARRI Reveal naturalistic color, no glamour, no oversaturation. Vertical composition, 9:16. negative prompt: blue hour, dusk, twilight, daylight, bright sky, sunset glow, drizzle, light rain, dry pavement, moving cars, parked cars, traffic, people, crowds, wide alley, oversaturated, HDR, neon overload, cyberpunk, teal-orange grade, CGI, 3D render, game-engine look, cartoon, lens flare, watermark, text artifacts

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

1

Source images and video this run consumed.

OIP.jpg

node 4 · 2.0 KB