I am really excited and proud to share the release of CommonSim-1 (@CSM_ai), a neural simulator that can consume text, images, and videos to create rich dynamic environments. (0/n)
This is the single best read on World Models and one of the most important reads in AI.
$10B has flowed into "world models" in the last 18mos, from Yann LeCun to FeiFei Li. The promise is, like LLMs, world models will provide the data it takes to scale robotics foundation models, and solve robotics.
..but the word has been abused to mean one of many things.
This post unpacks:
– What 5 traits makes a world model?
– How do the different approaches stack up?
– What is it used for within and beyond robotics?
– Where is the opportunity?
– Citations to research, news and blog posts
Companies / products in the space include:
– BigCo products: Google Genie, Tesla Optimus, Nvidia DreamDojo, DreamZero, Microsoft Muse
– Pure world model: AMI Labs, World Labs, Runway, Rhoda, Decart, Spaitial, Odyssey, Embo, Dream Labs, OneWorld
– Robot foundation model cos: Skild, Physical Intelligence, Figure, Mind
Very likely one of the seminal technologies of the next decade.
🚀Echo-2 is here - our new world model!
These aren’t videos. These are 𝟑𝐃 𝐬𝐜𝐞𝐧𝐞𝐬. Generated from a single image.
- Stunning visual quality.
- Real-time rendering.
- Interactive camera control.
- Physically grounded.
🧵More details👇
By popular demand, we're sharing more examples that showcase CSM Cube's ability to generate high-quality 3D mesh topology and distinct parts. The results shown were achieved using our single-click Image-to-3D feature, but users can obtain even better quality with careful prompting and editing.
CSM Cube delivers industry-leading model quality, topology, parts, and AI re-topology, providing a significant advantage for workflows from quick prototypes to full production.
We are thrilled to release the next leap in art-grade 3D generative models. Our single-click model pipeline gives unprecedented mesh outputs, with mesh parts-based topology. It is available now for all Cube tiers to start for free.
☑️ Our multi-stage hierarchical AI models produce a fully assembled 3D mesh with adaptive poly-counts, providing the clean, separated topology you need.
✅ A parts-based approach enables high-resolution meshes.
🌐 Quad and Triangle mesh support.
Access now: https://t.co/syFkGvNNmh
[New 3D Superpowers for All]
All free users now get trial access to our 2M poly models before upgrading 🚀
Get a high-resolution mesh with nearly 2 million polygons (no textures), plus a decimated version optimized for UV mapping with textures. Enjoy!
3D AI is reaching an inflection point. Feedback from a power user (EpicPhail - discord)
> "The new Kit-to-3D mode is capable of some truly insane stuff when combined with the increased max-poly count!"
> "Astounding how much time this can save indie devs."
> "All-in-all I think this took just over an hour of work to put together."
@CSM_ai 's new kit option is another big step forward in the world of 3D A.I.
Paired with the right Blender add-ons you can make an animation like this in 20 minutes or so.
#b3d#3D#AI#3dai#blender
We’re thrilled to launch Image to Kit - turn a single image into an animation-ready, modular kit to produce unmatched mesh quality with parts.✨🎨
We leverage inference-time compute to reason, zoom, and predict 3D parts with arbitrary resolution. Available now for premium users on CSM AI - upgrade now!
Blog and link below ⬇️
Excited to be featured on Meta's AI blog! Meta highlights how our team at Common Sense Machines leverages their open-source Segment Anything Model (SAM) within our 3D AI pipeline, enabling fast generation of compositional, near production-ready 3D assets 🎨🚀.
This integration empowers experts to precisely control 3D generative AI outputs and significantly reduces the barrier to entry for professional-grade 3D content creation.
Link below 👇
We are thrilled to release Chat-to-3D on CSM AI, to enable unprecedented mesh quality and controllability. This entire character was prompted end-to-end using this release.
🛠️ Automatically generate parts for complex assets
🌐 Generate T-pose outputs from any input
🎨 Generate style variations in one-shot
Image generation is powered by gpt-image. Available to all users (creative pro+ members get highest resolution outputs).
Text-to-4D Worlds in Production Environments 🛠️
We are thrilled to release open source agentic tools that stitch together:
🧊 CSM AI for 3D assets
✨ Free & instant text-to-animation with auto-rigging
🧩 Blender MCP add-on
💬 Your favorite client — Cursor, Claude, Windsurf, or custom scripts
Built for 3D artists and devs who want speed and control without leaving their favorite toolchain.
GPT-4o Image Generation to Part-based 3D Characters with PBR, in under 10 minutes ⚡️
Workflow:
🎨 Prompt GPT-4o to get an image (e.g., "3D asset of a styled character with all parts laid on a sheet for image to 3D")
🧩 Use CSM AI's Part-based tool to generate parts and assemble in Blender.
🚀 Big milestone at Cube! Our first-ever PBR model is live — turn any image or text into a 3D model with Albedo, Roughness & Metallic maps. Try it out https://t.co/Pey6h1yI8a
🎮 Built for game devs, VFX, and real-time creators.
Email [email protected] for free credits!
🚀 Thrilled to announce our partnership with https://t.co/5dB32rQh0o, where our AI-driven 3D technology is revolutionizing e-commerce model creation!
In a recent interview, Nextech's CEO Evan Gappelberg highlighted how our collaboration is reducing their production costs 50% while dramatically improving scalability. 🤝
🤖 As he put it, "We're lowering our cost and increasing our ability to scale. Those two things go hand in hand because essentially, you can't get scale using humans - you need AI." (link in comments)
We're proud to be recognized as pioneers in the AI space as we expand beyond gaming into e-commerce, helping create what Evan describes as "the highest quality 3D models on the market today."
Excited to continue building together and transformation 3D asset creation 🔥
#3dmodeling #ecommerce #ArtificialIntelligence
People might be sleeping on how well @cursor_ai with Sonnet 3.7 works for scripting simple games in @godotengine. Just open a local project in your IDE, and you can seamlessly jump between vibe coding and vibe 3D world-building. 🤯
The village blockout was created with a single prompt and is now easily customizable in the game engine editor. With @CSM_ai it is super easy to generate and add in more creative 3D models - like a custom gun, cauliflower gremlins, or background assets such as this shiny dragon.