maru retweeted
We're open-sourcing HyperFlow Infinity, the second release in the HyperFlow series.
HyperFlow Infinity achieves unlimited-length video generation: a minimal agent harness where HyperFlow generates a shot, watches it, updates a memory, and directs the next—in a recursive loop.
Our first film, "Alice in Wonderland: An Endless Wander", runs 60 minutes in a single pass, generated by Hyperflow Infinity, without extra model.
Code: github.com/Video-Rebirth/hyp…
maru retweeted
Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model.
🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal.
🔹 Full-project coding: games in Unreal, Unity & Godot.
🔹 Excels at parametric 3D modeling in Blender & CAD.
🔹 More reliable at finance work: Accurate modeling with hallucination down by 65%.
🔹 Matching the strongest proprietary models across coding, agentic and knowledge benchmarks.
Kimi K3.1 Pro and K3.1 Pro Highspeed (2–3× faster) are now live on Kimi.ai, Kimi Work, Kimi Code, and the Kimi API.
Open weights by Nov 4, 2026.
🔗 API: platform.kimi.ai
🔗 Tech blog: kimi.ai/blog/kimi-k3-1-pro
maru retweeted
Atlas lets you build out large 3D environments starting from a single input image.
You can then define a camera path through the world and generate a video along it.
Here’s how the 3D structure, camera path and final flythrough connect.
maru retweeted
if you look closely at the reliability test later in the video,
you’ll see a few examples of “physical reasoning” -- it tries something, it doesn’t fit (because physics), so it backtracks, switches gears, then does something entirely different to reach the goal.
in this case, it’s taking out the stuck puck, making space elsewhere, and putting it back in.
things like this (… a sort of physical problem-solving with step-by-step “chain-of-actions”) should just work with models out of the box.
How many pucks could a puck bot pack if a puck bot could pack pucks? 🏒
12 hockey pucks is a tight fit in this container – watch the model fix the arrangement when pucks get stuck or misplaced.
GEN-1.5 doesn't just puck around. Read more about it in our blog post and sign up for the waitlist in the comments below.
maru retweeted
How many pucks could a puck bot pack if a puck bot could pack pucks? 🏒
12 hockey pucks is a tight fit in this container – watch the model fix the arrangement when pucks get stuck or misplaced.
GEN-1.5 doesn't just puck around. Read more about it in our blog post and sign up for the waitlist in the comments below.
maru retweeted
🚀 Excited to introduce StreamRender-H3 — bringing real-time interactive worlds to life with diffusion models!
Built on Sol-Engine, we post-trained @MiniMax_AI H3 Ref2VA into a real-time streaming neural renderer.
🎮 Code drives the world. A game engine handles physics, object states, and player interactions, while H3 generates the visuals.
⚡ Real-time streaming: 2 denoising steps, 24 FPS, and ~273 ms per chunk on 8×GB200 GPUs.
🎨 One world, endless appearances: Change the reference image and prompt to render the same interactive world in entirely different styles.
🔥 Powered by streaming post-training, distillation, a fast causal H3 VAE, and full-stack inference acceleration.
Check out our demo and code! 👇
demo: nvlabs.github.io/Sana/Sol-En…
code: github.com/NVlabs/Sana/tree/…
maru retweeted
Qwen-Image-2.1 Consistent Lighting LoRA
blends seamlessly object into the scene
- no reflections/glares
- enhance specular highlights
- adds natural shadows
huggingface.co/RunningHubAI/…
maru retweeted
Retro anime sequence made with Scenario MCP:
>Opus 5.5 blocks every shot in Blender
>GPT Image 2.5 designs the pilot and the mech
>Seedance 2.5 animates it, ElevenLabs scores it
maru retweeted
Excited to introduce JEPA-Anything!
One world model that works across molecules, cells, fluids, patients, robots, or even weather.
Let's World Model Everything!
📄Paper: arxiv.org/abs/2609.20800
💻Code: github.com/Gen-Verse/JEPA-An…
maru retweeted
3D AI + Blender Gives Full Scene Control for Video Generation
Your AI video shouldn’t be a guessing game.
With 3D AI + Blender, you can build a scene, add AI-generated assets, and set the layout, camera, and composition before generating a single frame.
MCP and API workflows speed up the setup.
Less trial and error. More creative control.
maru retweeted
卧槽。这么小的模型，居然这么牛逼。。。
蚂蚁百灵前几天开源了一个设计模型，叫 Ming-Image-0.1-Design，才 6B。放在现在的生图模型里，基本属于小不点了。
然后这个小不点，在 Artificial Analysis 的 UI/UX 设计榜上，拿了开源第一。
我寻思了一下，以后生图，是不是就可以不花钱了？
它到底能干啥。
你说一句：画一台废弃工厂里的人形机器人，胸口发光，旁边飘个无人机。它啪一下就给你画出来，细节清楚，光影很有电影感。
更骚的是，它还有个兄弟，叫 Layer。
把这张图丢给 Layer，它能把机器人、无人机、背景一层层拆开，全是透明底，连被挡住的背景都自动补好了。想挪就挪，想换就换，跟在 PS 里打开分层文件差不多。
我是真的愣了一下。
先看视频我自己跑出来的👇
