Most people use @floraai for prompts. I use it to extract style.

Feed it a moodboard, a brand mark, a photo. It pulls the DNA palette, texture, mood and every image after sits in the same world.

That’s the difference between 40 random outputs and a style.

Video 3 in the FLORA @Contra University series is up.

Excellent Brother

Seedance 2.0 continues to surprise me — and I’m still actively testing its limits. As an experiment, I created a dynamic concept: a ninja being relentlessly chased by a futuristic Medusa — blending cinematic storytelling, fast-paced action, and visually striking environments. What I’ve discovered through multiple iterations: if you build the right narrative structure, Seedance 2.0 can intuitively reconstruct your original vision with impressive accuracy. Structure that works: • Location — define exactly where the scene takes place (city, desert, futuristic hall, rainy night street) • Emotion — set the tone and feeling (tension, euphoria, dark determination, adrenaline) • Action — describe step-by-step what is happening (who does what, and in what sequence) • Atmosphere — establish the overall visual style (cinematic, neon, raw, minimalistic) The more precise your structure, the better AI understands your intent — leaving less room for randomness. This unlocks a new level of possibilities for brands: • rapid validation of creative concepts • high-quality content at scale • scroll-stopping visuals for TikTok & Reels Adding sound design through ElevenLabs takes it even further — giving these visuals a whole new layer of immersion and realism. We’re entering a phase where creativity is no longer limited by production — only by imagination. If you're a brand, agency, or creator looking to explore AI-driven video content — feel free to reach out via DM. Open to collaborations and custom projects. Currently available on Higgsfield AI & ElevenLabs.

Haha, this could be a Mobile game Ad for Sure