• AI Valley
  • Posts
  • World Labs unveils multimodal world model

World Labs unveils multimodal world model

PLUS: OpenAI releases GPT-5.1

Author

Barsee
13 Nov

Together with

Howdy, it’s Barsee.

Happy Thursday, AI family, and welcome to another AI Valley edition. This issue takes 4 minutes to read.

Today’s climb through the Valley reveals:

  • OpenAI releases GPT-5.1 to all ChatGPT users

  • World Labs launches Marble, a multimodal world model

  • Plus trending AI tools, posts, and resources

Let’s dive into the Valley of AI…

REMIO 2.0

Image Credit: Remio

Tired of scattered notes, lost files, and forgotten conversations? 

Remio unifies your digital life, transforming your knowledge into a personal intelligence source.

Here’s what you’ll love about remio:

  • Tailored Answers: remio provides unique answers by combining with your personal knowledge.

  • No More Uploads to ChatGPTs: One click to sync all your files, making your entire knowledge base chatable with AI.

  • Master Your Meeting: Unlimited free recording with transcription, get AI summaries with key decisions.

  • Privacy & Security: With a "Local First" design, all your data is stored exclusively on your device.

*This is sponsored

THROUGH THE VALLEY

Source: OpenAI

OpenAI has introduced two new models called GPT 5.1 Instant and GPT 5.1 Thinking. Both are designed to improve tone, speed, and user control. Instant feels warmer, more conversational, and better at following instructions. Thinking is quicker on simple tasks and more persistent on complex ones, adjusting how long it thinks based on difficulty.

GPT‑5.1 introduces adaptive reasoning, meaning the model decides whether to respond instantly or take extra time for complex tasks to generate more thoughtful and accurate answers. This is reflected in substantial improvements on math and coding benchmarks, showing genuine gains in practical reasoning and output quality.

Source: OpenAI

Users can now choose from eight tone presets: Default, Professional, Friendly, Candid, Quirky, Efficient, Nerdy, and Cynical. You can also customize emoji usage, warmth, and conciseness through new personalization settings. These preferences apply across all chats instantly.

GPT‑5.1 Instant and Thinking begin rolling out today, starting with paid (Pro, Plus, Go, Business) users and then to free and logged-out users in the coming weeks, while GPT-5 stays available for three months under legacy options.

Why does it matter?

OpenAI appears to be shifting its strategy from chasing pure intelligence to creating deeply personalized and always available assistants. By allowing users to define how ChatGPT sounds and behaves, the company is moving away from the idea of one model for everyone and toward AI that feels individually tailored to each person.

Source: World Labs

World Labs has released Marble, a world model that turns text, images, and videos into complete and editable 3D environments. It is one of the clearest moves toward AI that understands physical space instead of only processing language.

Marble can generate persistent 3D worlds from almost anything, such as text prompts, photos, panoramas, video clips, or rough layout sketches. Creators can also design simple room shapes or object placements using Chisel, then style the entire environment with text prompts. Structure and appearance stay separate, so you can revise one without breaking the other. This makes iteration far easier than previous approaches.

The model is available now with free and paid plans beginning at twenty dollars per month. Early uses include game development, VFX, architecture, robotics simulation, education, and VR prototyping.

Why does it matter?

This release arrives during a global push to solve a deeper limitation in AI. Modern models are fluent with language, yet extremely weak at the world. They struggle with physics, causality, and state. This is why autonomous cars still fail on edge cases and why industrial robots remain fragile in unpredictable settings. As one researcher noted, current AI can describe the world, but it cannot truly grasp it.

World Labs CEO Fei Fei Li refers to this missing ability as spatial intelligence. The goal is to ground AI in the geometry, physics, and dynamics of real environments. World models attempt to bridge this gap by predicting what might happen next, not just what a scene looks like in the moment. They update objects and relationships when something changes. If an object moves or falls, the internal world adapts.

This is the core difference. Video models focus on appearance. World models simulate reality. They are interactive, reactive, and capable of maintaining a long term internal state.

TRENDING TOOLS

  • Build with MongoDB > From prototype to production. Give your AI agents long-term memory and context *

  • GPT-5.1 > OpenAI’s upgraded model with full personality customization

  • Hyperlink > The first AI super-assistant that lives directly inside your computer

  • Marble > Turn a single image, video, or text prompt into a detailed 3D world you can explore and interact with

  • Fuser > A unified creative workspace built for every AI model and every type of media

  • Video Localization by Algebras > Human-like cultural dubbing across 32 languages

  • Kadabra > Automate your workflows in minutes using AI

  • Gemini Live > Now speaks faster, uses accents, and sounds more expressive, perfect for quick lecture reviews or real-time language practice

(*) signifies sponsored tool

THINK PIECES / BRAIN BOOST

THE VALLEY GEMS

What’s trending on social today:

THAT’S ALL FOR TODAY

Thank you for reading today’s edition. That’s all for today’s issue.

💡 Help me get better and suggest new ideas at [email protected] or @heyBarsee

👍️ New reader? Subscribe here

Thanks for being here.

REACH 100K+ READERS

Acquire new customers and drive revenue by partnering with us

Sponsor AI Valley and reach over 100,000+ entrepreneurs, founders, software engineers, investors, etc.

If you’re interested in sponsoring us, email [email protected] with the subject “AI Valley Ads”.