Tugg Speedman

you m..m..make me happy *retard seal claps*

[RepoCatAI | Sharing GitHub Projects for AI & Robot @repocatai_git]

【Mind-Blowing】
An AI that generates “walkable 3D worlds” from a single image or text is here🔥
Matrix-3D (built with Python), ⭐️~800.

Here’s what it can do👇
・Generate vast 3D scenes from images/text
・Panoramic spaces you can explore in 360 degrees
・Freely specify movement trajectories
・A memory-efficient version that runs on 12GB VRAM is also available

For anyone who wants to create game backgrounds or VR spaces with AI.
“Generating entire worlds” is now a reality.

https://t.co/nO3iujyH5P

[小互 @xiaohu]

3D 圈的人看到这个应该会沉默几秒

3D 模型再次进化

Hyper3D 这个 BANG to Parts功能有点牛P

你丢给它一个完整的 3D 模型，AI 会自动帮你把它拆解成一个个独立的、完整的的3D零件

也就是可以将任何完整的 3D 模型自动拆解成多个3D的零件

拆好的零件可以直接用来做动画、改材质、甚至去 3D 打印

这能帮 3D 设计师省下大把“手动擦屁股（清理模型）”的时间

更狠的是：拆完还能接着拆

整车 → 拆出车轮 → 车轮再拆出轮毂、轮胎、螺丝。

理论上你想拆多细拆多细。

而且每个零件都是独立完整的 3D 模型：单独换材质、单独改形状、甚至单独重新生成一个新的换上去

3D 模型第一次变成了“乐高”，而不只是“手办”

还有一个关键点是：拆出来的每个零件都是完整、独立、可编辑的 3D 模型。换材质、改造型、缩放变形、甚至整个重新生成一个换上去,都行

借助目前Codex和Claude code的强大能力，打造一个 3D 模型/开发工作流，非常牛P

[am.will @LLMJunky]

WOW! I thought Opus 5 would just be a cheaper Fable. I was dead wrong.

This model is absolutely BONKERS. I'm genuinely blown away.

Opus 5 built the best Rocket League clone i've ever seen, and it did it while consuming only 27% of my 5x Max sub.

Sound on. There's layers to this onion. I'm floored rn.

At least for building games and 3D modeling, this has to be the best model in the world by a notable margin.

Look at the car. The reflections. The details.

Playable game & open source code in the comments.

[Leon Derczynski ⚒️☁️🏔️🌲 @LeonDerczynski]

I keep saying the strength is in the harness, not the model - because it's true. Not much use without a harness, though.

Here's VISA's open-source cybersecurity harness. Just add model; use hosted closed models, use hosted open models, run on-prem, whatever. Very cool of them to share this tech and lift the defensive cybersec poverty line.

[Sen Zh @SenZh4]

Vibe Coding 能把简历玩出什么花样？？

😄分享一个最近Vibe的个人网站，用AI把自己变成一个3D虚拟小人，还Vibe了一个小白也能快速生成的平台。#vibecoding #3d互动 #网页设计

[Okan Can @0kncn]

Bir süredir Spider-Verse havasında bir şehir salınımı denemek istiyordum. Karakter için tek bir görsel kullandım, hareketi de 9 karelik depth storyboard’la kurdum.

Prompt 🔻

[Abyssal @abyssallD]

MOST PEOPLE STILL DESCRIBE CAMERA MOVES IN TEXT AND HOPE THE AI UNDERSTANDS

Eyecandy gives you real cinematic techniques as GIFs and clips you can download and drop straight into Seedance as video references.

> Dolly zooms
> Bullet time
> FPV
> Dream state
> Double dolly — all already filmed and ready.

Visual technique library: https://t.co/Gqxg8OBvC5

[Spencer Sterling @cerspense]

I don't think i can go back to regular software anymore

exploring a crazy combo of real-time depth estimation, raymarched SDFs, StreamDiffusion and BiRefNet. The entire setup was authored by GPT 5.6 Sol and Luna in Sentinel over MCP while I steered it in real-time.

[しとちゃ！🦊🍮 @nemumusitocha]

Opus５でもBlenderMCPのお城作りためしてもらったなんだけど、Fableと比べてもレベル違いすぎて笑ったんｗｗｗｗなんで同バージョンの最上位モデルより強いのだｗｗｗ

[Avi Chawla @_avichawla]

The next step after Karpathy's autoresearch idea:

Tuning an agent is mostly manual work, done by editing prompts, tools, and control flow by hand and rerunning the evals to see what moved.

Researchers have been building several automated optimizers to do that outer loop instead.

The underlying process is the same:

- An LLM proposes a change
- An evaluator scores it
- And the proposer reads that result before proposing the next one.

They differ in what they edit and what feedback they get to read.

1) Berkeley built GEPA that optimizes the text of a system, like prompts, tool descriptions, or the agent's own code.

Instead of collapsing a run into one scalar reward the way RL does, it reads the full execution trace (errors, reasoning, tool output), diagnoses why the run failed, and proposes a targeted fix.

It also keeps every candidate that's best at some part of the task, not just the one with the highest average score. So a strong specialist survives even when a more balanced candidate beats it overall.

That trace-level feedback lets it converge in hundreds of rollouts instead of the thousands that GRPO needs.

2) AutoResearch, inspired by Karpathy, runs a narrower version of the loop, where a coding agent iterates on a program(.)md file, scores the outputs against fixed evals, and keeps whatever improves.

3) Meta-Harness points the loop at the harness itself, the scaffolding code that decides what to retrieve, how to format it, and what state to keep between calls.

The natural question is which of the three to use. And the answer is none of them alone.

On the Frontier-CS benchmark, holding the model, thinking effort, and budget fixed, no optimizer wins everywhere. Across 10 tasks, GEPA led on 3, AutoResearch on 3, and Meta-Harness on 4.

Each optimizer hill-climbs fast and then stalls, making most of its progress in a few iterations before flattening out.

But the researchers found that handing the stalled candidate to a different optimizer breaks the plateau, since each one attacks the problem differently.

To use this in practice, omni (open-source) already automates all of that.

It runs every optimizer on a fraction of the budget, takes the best candidate, and hands it to a fresh optimizer to keep going.

It scores 7.8 percentage points above the best standalone optimizer at the same budget, and finishes faster.

The whole meta-optimizer is around ten lines in the optimize_anything API, the same interface these optimizers already run through.

You can also point an agent at the gepa-ai/gepa repo and have it use the gepa-optimize-anything skill.

Here's the repo: https://t.co/NftJXozw04

For a first-principles guide to GEPA and the engine under most of this, my co-founder wrote a full breakdown of why reflective evolution beats GRPO by around 10 points with 35x fewer rollouts and no GPU training, purely from natural-language reflection on trajectories rather than backprop on weights.

Read it below.

[Chong-U @chongdashu]

Opus 5 with Unity CLI feels like a cheat code for vibe coding games with AI.

Used it to spruce up a scene Fable 5 built to give it a distinct style.
→ Stylised water
→ Wind foliage
→ Scene lighting + contact shadows
→ Depth outlines

Its speed makes it faster to iterate, while still being intelligent enough to handle tasks.

[3Dgamedev @unity3dvfx]

This is crazy

Claude + Unreal Engine MCP creating 3D building from a single prompt 🤯

[Nes @nesdesignco]

No game engine. No 3D skills. Every object made with @MeshyAI
Every line written by @AnthropicAI Claude Fable.
Rendered in @threejs (WebGPU).
An explorable candlelit alchemist's chamber of mushrooms in your browser. 🍄 https://t.co/3KVkdkAb4F
https://t.co/TPfABOq3Nt

[mikuru @_mikuruwu]

fun fact, I created a cel shaded look in After Effects for this shot with normal pass!

#aftereffects

[Justine Moore @venturetwins]

FLUX 3 is insanely good at one-shot timelapses 🤯

In this case, I gave it a single photo of a venue and asked it to work backwards - "generate a continuous timelapse of event staff setting this up from an empty room."

This was the first try.

[Jahnavi Kapoor @JanhvikapoorFC_]

Tap and Hold.....!
Definately you are Blessed 🌸

[cheaty @cheatyyyy]

Now that I can talk about FLUX 3, it is an amazing model!

It has the best native audio out of any video gen by far, and it rivals Seedance 2, generates videos upto 20s long, supports image/video/audio input and supports upto 10 of them! You can mix and match your references!

[GMI Cloud @gmi_cloud]

Crazy. Opus 5 outperformed Fable 5 on 3d modeling at 2/3 the price

We tested Opus 5, Fable 5, and GPT 5.6 Sol on the same prompt: build an explorable Japanese suburban street in Three.js, full 3D, rendered like a hand-drawn anime background. All on Max effort.

Opus 5 took the longest but delivered the most varied scenery: clean dark outlines, flat cel-shaded color bands, dense street details down to the vending machine signage.

Fable paid more attention to details, and while GPT 5.6 Sol was 3-4x cheaper and the fastest, but the render shows it: sparse, washed out, barely styled.

Opus 5: 1hr 21min · $26.7
Fable 5: 1hr 5min · $35
GPT 5.6 Sol: 41min · $7.1

[Spectro 🇺🇸 @Spectromachina]

I started using Blender through MCP about two weeks ago, and I quickly realized that you can build almost anything with AI.

This model was created using Blender, Hunyuan3D, Gemini, and ChatGPT.

Here’s how I did it:

I opened Gemini, uploaded an image of the Gundam model, and asked it to generate a clean front-view image.

I uploaded that front view to ChatGPT and asked it to generate two additional angles: a back view and a 45-degree front-left view.

I went to:

https://t.co/H3cfw2AUE9

I signed up with my email and translated the page into English. Then I opened Image to 3D and selected the multi-image option.

You’ll see a diagram of a whale from several angles. Upload each reference image in its corresponding position, such as front, 45-degree front-left, and back.

I selected the 1.5M-face option. This produces a very high-poly model, but don’t worry, we’ll fix that next. Once the generation is complete, download the model as a GLB file.

From the Hunyuan homepage, open 3D Studio using one of the dropdown menus. Select the topology or retopology tool and upload your GLB.

I chose the High setting to preserve as much detail as possible. After a few seconds, the model was retopologized. It kept most of its visual detail while using far fewer polygons.

The original head and rifle didn’t look very good, so I generated them separately.

I returned to ChatGPT and created dedicated reference images for the Gundam’s head and rifle. I generated each part individually in Hunyuan at the 1.5M setting, then ran both through the same retopology process.

Next came the textures.

Open the texture section, select the multi-image option, and upload the same reference images according to the whale orientation indicators.

However, instead of using the original clean textures, I asked ChatGPT to recreate them with wear, rust stains, scratches, and other surface damage. This gave the Gundam a much older and more authentic appearance.

Once every part was textured, I imported everything into Blender.

You can ask Codex through MCP to remove the original head and rifle, or you can do it manually.

Select the main model and press Tab to enter Edit Mode. Press 3 to enable face selection, then press C to activate Circle Select. Paint over the faces belonging to the original helmet or rifle.

You can press X to delete those faces or P to separate them into another object. Then position the newly generated, more detailed head and rifle in their place. I demonstrate this process in one of my older videos.

And that’s it. You now have a very cool 3D Gundam model!

Afterward, I created the cockpit, separated the model into movable sections, and rigged everything for use in my Three.js game. That process deserves its own tutorial, though.

If anyone wants to see it, let me know.

Or just ask your AI, I guess. They seem to know everything these days. xD

[Jhosep Chevarria @Jhosep_gamer]

Evolution of my custom global illumination plugin for UE5, no baking, no engine mods, no DXR (Hardware raytracing) 100% realtime, vs Lumen. Work in progress, having a lot of fun. #UE5