World Connexion

IA maker cg artiste Unity Game
Hypnose

[烁皓 @eternityspring]

本地能跑 MiniMax H3 之后，都以为 AI 短剧的门槛天塌了

真正去做才发现，要把小说改成剧本然后分镜制作就必须得先固定人物角色。

我开源了一个 skill 专门来做这件事，Claude Code / Codex 都能跑：
- 丢一本小说进去，AI 自动拆出每个角色：人物画像 + 卡通形象 prompt + 音色 prompt
- 如果你装的有 codex cli 环境，还会调用 Codex 生成角色三视图
- 一条命令 /novel-characters ./你的小说.txt 跑完

走你当前订阅的额度，不用另配 API key。最终会生成一个 report 页面以供审阅

项目地址：https://t.co/AwV9LsQBSH

欢迎感兴趣的朋友评论区交流。

[Fouad | فؤاد @Bp__Dev]

I'M MAKING A GAME I ALWAYS WANTED TO PLAY

[MeshyAI @MeshyAI]

📢 Multi-View Image to 3D just got a major upgrade!

Now your textures generate from multiple views too.

So a few references of the same object bring the whole model back with much higher accuracy and consistency, from structure to texture.

[Kfir Aberman @AbermanKfir]

1/ We’re entering the era of agentic commerce, and we believe that world models are the missing engine.

Instead of browsing static product pages, agents will let you interact with products inside your own world before making decisions.

Today, @DecartAI we're releasing Anywear👇

[takkun @takkun3d]

平面のポリゴンを立体的な部屋にする「Parallax Cube Mapping」ツールが形になりました！

最高に可愛い部屋のイラストは @eringi_inochi さんに描いていただきました！大感謝！🙇‍♂️

UIからサクッと適用できて、ビューポートでのプレビューとArnoldでのレンダリング両方にバッチリ対応してます💪

[TAngraFX @TAngraFX]

MotionGen can now simulate Slow motion and fast motion in the same video clip. Full tutorial on the TAngraFX YouTube channel.
You need at least V2.9.4 with engine v1.4.6
100% local, one-time payment. Lifetime use.
No internet
Auto animation enhancement.
Auto foot lock

[World Connexion @Klr_Kilik]

#recette #tiktok #lotus

[Massimo @Rainmaker1973]

Watch this brilliant 3D printed fish tank skeleton dancer.

[📹 arlo_finds]

[Two Minute Papers @twominutepapers]

Incredible AI parkour simulation from @NVIDIAAI. It learned to be a pro from looking at just 30 seconds of footage! ▶️Full video: https://t.co/LPD9rNjtJQ

[World Connexion @Klr_Kilik]

🎮 Joy Impact: The Bass Wall for Beat Gamers
A high-energy VR rhythm experience where you smash the bass wall and feel every beat.
Meta Quest → https://t.co/ENoUEusDEn
#gaming #indiegames #MetaQuest #VR

[John @johnAGI168]

INS 爆款巨人之手娃娃穿搭👚

怎么感觉有点解压😄

Seedance 2.0 prompt 👇

【风格】巨人之手娃娃穿搭（GWRM · Giant Hands Barbie Doll Styling），真实手机竖屏实拍质感（9:16 Vertical, Photorealistic），ins爆款粉色少女感，自然家居光
【时长】15秒
【场景】真实居家客厅：浅色木地板、白色衣柜、墙上挂着电视，自然窗光从侧面打进来；女主以真人芭比娃娃的比例站在地板上，画外伸进来的手比她整个人还大
【角色】女主@图片1（被缩小成娃娃比例的真人，面部与身材保持真人质感，动作有轻微娃娃感的僵直和顿��）
【核心机制】比例差：一双巨大的手（正常人手，相对女主如巨人）从画外反���伸入画面为她换装——捏起、放下、脱衣、套衣、蹲下换鞋；女主全程配合但保持娃娃式的乖巧站姿，被摆弄时身体轻微晃动
【常驻字幕】画面上方粉色手写花体英文标题"GWRM"，下方小字"Styled by Giant Hands Like a Barbie Doll"，配粉色蝴蝶结与星星装饰

[00:00-00:02] 开场：把娃娃放下（Placement）
巨手捏住女主的腰部把她从画外拎进来，轻轻放到木地板上站稳，她穿着粉格睡衣外套、白色吊带、白色蕾丝短裤和白袜，头上戴粉色睡眠眼罩，双脚落地时轻轻晃了一下才站定。

[00:02-00:05] 第一步：摘眼罩、脱外套（Undress）
巨手两指捏住她头上的粉色眼罩向上一摘；随后双手捏住粉格外套的两只袖口，把整件外套从她身上向后一褪、整件拎出画面，她只剩白吊带和蕾丝短裤，抬手揉了揉眼睛。

[00:05-00:08] 第二步：护肤上妆（Skincare Close-up）
切怼脸特写：巨大的手指蘸着乳白色面霜在她脸颊上轻轻抹开，她乖乖闭眼仰着脸；接着一根手指举起一面粉色小镜子递到她面前，她睁眼照镜子、抿嘴整理刘海。

[00:08-00:11] 第三步：换新装（Dress Up）
切回全景：巨手捧着一件绿色荷叶边抹胸上衣从她头顶套下、拉正下摆；再取一条粉白格子超短裙从她脚下向上提到腰间，抚平褶皱，她配合地抬起手臂又放下。

[00:11-00:13.5] 第四步：换鞋袜（Shoes & Socks）
低角度贴地特写：巨手捏着粉色长筒蕾丝袜一只只往她���上套、拉到膝盖；再把一双玫红色尖头高跟鞋摆到她脚边，捏着她的脚踝一只只塞进鞋里，指尖调整鞋跟站稳。

[00:13.5-00:15] 收尾：走秀定格（Final Pose）
巨手退出画面，女主独自站在地板中央，双手叉腰扭胯摆pose，接着原地转身甩了下头发，对镜头歪头一笑，画面定格在她的全身穿搭上。

【音效】轻快甜美的ins风BGM打底，衣物摩擦的窸窣声、鞋跟磕地的"哒"声、手指擦面霜的细微声都保留真实录音质感，每次换装完成配一声轻"叮"，结尾定格配上扬收束音

[Omma AI @omma_ai]

This is real-time. What a time to be alive!

By @marco_fugaro made with Omma.

[3DxDEV @3DxDEV7]

This Procedural Clouds Shader drops realistic cloud layers into your scene — and it works in both Cycles AND EEVEE.
What makes it great is that it's fully procedural, so there are no external textures to manage. It's lightweight, fast, and easy to tweak.

Read More : 🔗 https://t.co/nwQLA78VWA

[ZEUS⚡️ @zeuuss_01]

OPUS 5 + HIGGSFIELD JUST REBUILT A 2005 PS2 ADVENTURE GAME WITH 2026 GRAPHICS

no engine. no studio. no art team. what if the adventure games i grew up on shipped today?

this is what came back.

00:00 - drop straight into the fight, blades out, forest lit with god rays
00:03 - golden bow - one arrow, a creature bursts into leaves
00:07 - the ground splits open, a moss-stone golem boss rises, health bar up top
00:11 - dodge the fist slam, unload on the weak point, HP + mana bars live

third-person. real HUD. real game-feel.

custom hero, custom boss, every frame generated, nothing hand-modeled.

no build, no assets - one afternoon and a text prompt.

this is where gamedev is heading.

follow + reply "REBUILD" if this hit you.

the exact prompt i used ↓

[angel @angeldot_]

CLAUDE OPUS 5 HA RECREADO UN JUEGO DE POKÉMON EN 3D

Y no es un vídeo, es una web jugable.

Corre en el navegador, sin descargas, sin instalar nada.

Web: https://t.co/1tf3fRBYEU

Empiezas en Pueblo Paleta, te mueves por el pueblo, entras en las casas hasta puedes elegir inicial.

Lo lejos que ha llegado el vibe coding es una locura.

[Casey Sheep @casey_sheep]

The new Blender VFX Add-on is currently in development. We're planning to release it sometime next week. It's still a work in progress, so stay tuned!
#b3d

[World Connexion @Klr_Kilik]

https://t.co/SvjpKazyJu

[Madni Aghadi @hey_madni]

SOMEONE VIBE CODED A FULL PC GAME

100% prompts with no code

the stack:

- built entirely in Cursor
- 90% Grok 4.5, rest Opus 4.8 and GPT 5.5
- switched to Grok 4.5 for speed and cost
- built on Godot
- art generated with ChatGPT Images through his own custom tool

Act 1 is already playable

this is what vibe coding actually looks like now

[Abhishek @HeyAbhishek]

AI anime production is getting powerful.

GPT Image 2 created the entire storyboard from one prompt.

Then Seedance 2.0 turned it into a cinematic anime scene.

Step-by-step tutorial with prompts: 👇

[MeshyAI @MeshyAI]

🚀 Smart Topology is LIVE in Image-to-3D!

Generate 3D assets with native part segmentation, clean and accurate, in ~10 seconds. Full control over poly count.

Made for game dev workflows, real-time and web-based 3D.

We'd love your feedback to help us refine the quality!