FluxmaniaV + Nolan LoRA | Veo 3.1!

Evaluación mensual de modelos entrenados por la comunidad, probados en ComfyUI para medir su fidelidad de estilo, respuesta al prompt y consistencia visual. Esta edición analiza FluxmaniaV y el Cinematic Film Stock LoRA (v1.0), explorando su comportamiento bajo prompts controlados y condiciones reproducibles.

MODEL SETUP

checkpoint

(loading)

workflow

download

LoRA #1

(loading)

LoRA #2

(loading)

PROMPT [+]

COPIED

PROMPT [−]

COPIED

(loading)

WORKFLOW

(loading)

(loading)

PROMPT [+]

COPIED

PROMPT [−]

COPIED

Prompt A

overtheshoulder

Front view medium shot neutral expression

Front view, medium shot, neutral expression

Medium Shot

Wide Shot

wide shot establishing scene

Close-Up

Three-quarter view, low angle, introspective mood

Wide Variant

wide shot variant 2

Low angle, dramatic lighting, tense expression.

Low angle dramatic lighting tense expression

Medium long shot, centered composition

Medium long shot centered composition cinematic frame

Front view, looking directly at camera

Front view looking directly at camera symmetrical frame

Over-the-shoulder shot, focus on tunnel ahead

Overtheshoulder shot focus on tunnel ahead

Front view medium shot neutral expression

Front view medium shot neutral expression

Front view medium shot neutral expression

Front view looking directly at camera symmetrical frame

Three-quarter view, low angle, introspective mood

Front view looking directly at camera symmetrical frame

5219 AI visuals character rendering Cinematic AI ComfyUI film style FluxmaniaV LoRA test Nolan LoRA prompt engineering