Latte's banner
Latte's profile picture

Latte

@0xbisc • 14,996 subscribers

Full Stack Designer, AI Explorer // Prompt & Workflow Sharing // Flora · Reve · Dreamina · TapNow · WaveSpeed Creative Partner ✽ Open for work

Shorts

Why is this fish so slippery? made with Seedance 2.5 on OpenArt

Why is this fish so slippery? made with Seedance 2.5 on OpenArt

257,420 views

It started with one dragonfly made with Seedance 2.5 on OpenArt

It started with one dragonfly made with Seedance 2.5 on OpenArt

35,472 views

Don’t interrupt my noodles made with Seedance 2.5 on OpenArt

Don’t interrupt my noodles made with Seedance 2.5 on OpenArt

54,400 views

Kitchen Hunt GPT Image 2 + Seedance 2.0 is the best combo right now for speeding up creative workflows. Prompt below ↓

Kitchen Hunt GPT Image 2 + Seedance 2.0 is the best combo right now for speeding up creative workflows. Prompt below ↓

308,909 views

The Noodle Master 🎬 Seedance 2.0 Prompt shared below

The Noodle Master 🎬 Seedance 2.0 Prompt shared below

369,165 views

Looks like they won’t be collecting that bounty made with Seedance 2.5 on OpenArt

Looks like they won’t be collecting that bounty made with Seedance 2.5 on OpenArt

23,326 views

never step on a fish made with Seedance 2 #DreaminaCPP Dreamina AI

never step on a fish made with Seedance 2 #DreaminaCPP Dreamina AI

69,517 views

Grace Dive made with GPT Image 2 + Seedance 2 #DreaminaCPP Dreamina AI Prompt below ↓

Grace Dive made with GPT Image 2 + Seedance 2 #DreaminaCPP Dreamina AI Prompt below ↓

186,792 views

Wing Chun Flow made with GPT Image 2 + Seedance 2 #DreaminaCPP Dreamina AI Prompt below ↓

Wing Chun Flow made with GPT Image 2 + Seedance 2 #DreaminaCPP Dreamina AI Prompt below ↓

182,429 views

Ink Groove made with GPT Image 2 + Seedance 2 #DreaminaCPP Dreamina AI Workflow and prompts below ↓

Ink Groove made with GPT Image 2 + Seedance 2 #DreaminaCPP Dreamina AI Workflow and prompts below ↓

192,896 views

kinda bored, just messing around with a pen made with Seedance 2 on OpenArt

kinda bored, just messing around with a pen made with Seedance 2 on OpenArt

62,878 views

TENNIS SHOWDOWN made with GPT Image 2 + Seedance 2 on OpenArt Prompt below ↓

TENNIS SHOWDOWN made with GPT Image 2 + Seedance 2 on OpenArt Prompt below ↓

126,635 views

I simulated a One Piece ARPG game. Started with a GPT Image 2 screenshot. Prompt: Generate an ARPG dialogue game screenshot inspired by [film/series name] Then extended it into gameplay with Seedance 2.

I simulated a One Piece ARPG game. Started with a GPT Image 2 screenshot. Prompt: Generate an ARPG dialogue game screenshot inspired by [film/series name] Then extended it into gameplay with Seedance 2.

133,018 views

The Arrival of Giants Seedance 2.0 Prompt shared below

The Arrival of Giants Seedance 2.0 Prompt shared below

106,484 views

Wrong cat, buddy Made with Seedance 2.0

Wrong cat, buddy Made with Seedance 2.0

127,457 views

The cat checks into the Grand Budapest Hotel Made with Seedance 2.0

The cat checks into the Grand Budapest Hotel Made with Seedance 2.0

122,945 views

THE OLD SKATE KING made with GPT Image 2 + Seedance 2 on OpenArt Prompt below ↓

THE OLD SKATE KING made with GPT Image 2 + Seedance 2 on OpenArt Prompt below ↓

71,928 views

This was a fun experiment. I took a different approach to make the coffee look more appetizing, and honestly, it made me want a cup while I was working on it. Create a hyper-realistic exploded vertical ingredient deconstruction infographic of a chocolate coffee drink. Centered vertical alignment, each component floating separately with clear spacing: cocoa powder particles (top), glossy dark chocolate sauce ribbons and droplets (separate layer), airy whipped cream mound (separate layer), smooth chocolate-coffee liquid layer (separate layer), roasted coffee beans (separate layer), sugar crystals (separate layer), and at the bottom an empty small transparent ribbed glass with a weighted base. Deep premium dark background: clean studio backdrop with a near-black graphite-to-charcoal gradient, subtle and smooth. Soft controlled studio lighting, crisp separation shadows under each layer, ultra-sharp macro realism, minimalist thin pointer lines and labels, high-end product infographic, 8K.

This was a fun experiment. I took a different approach to make the coffee look more appetizing, and honestly, it made me want a cup while I was working on it. Create a hyper-realistic exploded vertical ingredient deconstruction infographic of a chocolate coffee drink. Centered vertical alignment, each component floating separately with clear spacing: cocoa powder particles (top), glossy dark chocolate sauce ribbons and droplets (separate layer), airy whipped cream mound (separate layer), smooth chocolate-coffee liquid layer (separate layer), roasted coffee beans (separate layer), sugar crystals (separate layer), and at the bottom an empty small transparent ribbed glass with a weighted base. Deep premium dark background: clean studio backdrop with a near-black graphite-to-charcoal gradient, subtle and smooth. Soft controlled studio lighting, crisp separation shadows under each layer, ultra-sharp macro realism, minimalist thin pointer lines and labels, high-end product infographic, 8K.

133,405 views

Why Won’t It Stay?! 🎬 Seedance 2.0 Prompt shared below

Why Won’t It Stay?! 🎬 Seedance 2.0 Prompt shared below

94,288 views

Phew… that was close Made with Seedance 2.0

Phew… that was close Made with Seedance 2.0

107,836 views

Videos

LIVE
1.2k

Anya Rossi

sweetdream.ai

SweetDream.aiSponsored•Livecam
Streaming Now

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

HD live stream
Exclusive private shows
1.2k viewers online

Free preview available • Premium content

Before the Void Swallows You Made with Wan 3.0 30s in one go, the style stays consistent from start to finish, and the motion and overall coherence feel noticeably better. Also just noticed WAN 3.0 is ranked first on Artificial Analysis for both Video Editing (with Audio) and Text-to-Video (with Audio).
0:30
0xbisc's profile picture

Before the Void Swallows You Made with Wan 3.0 30s in one go, the style stays consistent from start to finish, and the motion and overall coherence feel noticeably better. Also just noticed WAN 3.0 is ranked first on Artificial Analysis for both Video Editing (with Audio) and Text-to-Video (with Audio).

Latte

13,711 views • 7 days ago

I tested Seedance 2.0 vs Kling 3.0. The winner was obvious. Same prompt, same reference image, one scene: a boy rescuing his dog from an oncoming car. Seedance 2.0 came out ahead across the board: · Better character consistency — the boy's face and expression stayed recognizable through fast, dynamic motion. · More coherent physics — when the boy dives to save the dog, he actually lands on the other side of the road. In Kling 3.0, they just drop in place, which breaks the logic of the scene. · More natural motion — fluid movements without the exaggerated, theatrical quality that plagues most AI video. · Richer cinematic language — varied shot compositions including close-ups, camera transitions, and framing shifts that actively reinforce emotion and narrative tension. The gap suggests Seedance 2.0's training leans heavier into film language, not just motion synthesis.
0:20
0xbisc's profile picture

I tested Seedance 2.0 vs Kling 3.0. The winner was obvious. Same prompt, same reference image, one scene: a boy rescuing his dog from an oncoming car. Seedance 2.0 came out ahead across the board: · Better character consistency — the boy's face and expression stayed recognizable through fast, dynamic motion. · More coherent physics — when the boy dives to save the dog, he actually lands on the other side of the road. In Kling 3.0, they just drop in place, which breaks the logic of the scene. · More natural motion — fluid movements without the exaggerated, theatrical quality that plagues most AI video. · Richer cinematic language — varied shot compositions including close-ups, camera transitions, and framing shifts that actively reinforce emotion and narrative tension. The gap suggests Seedance 2.0's training leans heavier into film language, not just motion synthesis.

Latte

167,558 views • 6 months ago

made this short tribute to Wong Kar Wai’s Fallen Angels with MiniMax H3 only used one prompt and two character reference images the native 2K output made a noticeable difference here, especially in the details and motion and the cost is much lower than Seedance 2 more experiments soon MiniMax Design (H3) #MiniMaxH3
0:29
0xbisc's profile picture

made this short tribute to Wong Kar Wai’s Fallen Angels with MiniMax H3 only used one prompt and two character reference images the native 2K output made a noticeable difference here, especially in the details and motion and the cost is much lower than Seedance 2 more experiments soon MiniMax Design (H3) #MiniMaxH3

Latte

20,925 views • 1 month ago

I made this soda commercial with invideo Agent One through a few simple conversations. Here’s the full workflow:
0:28
0xbisc's profile picture

I made this soda commercial with invideo Agent One through a few simple conversations. Here’s the full workflow:

Latte

21,247 views • 3 months ago

Youmind 这个功能太赞了！三步就能帮你把照片变成 Youtube 质感封面： 1. 找张你喜欢的参考图 2. 让 YouMind 生成风格提示词 3. 保存为快捷指令，以后一键生成 不止 Youtube，只要是你喜欢的风格都能转变成可复用模版
1:26
0xbisc's profile picture

Youmind 这个功能太赞了！三步就能帮你把照片变成 Youtube 质感封面： 1. 找张你喜欢的参考图 2. 让 YouMind 生成风格提示词 3. 保存为快捷指令，以后一键生成 不止 Youtube，只要是你喜欢的风格都能转变成可复用模版

Latte

21,498 views • 9 months ago

No more content to load