GENERATIVE AI

→ View original post on X — @runwayml

Runway and NVIDIA launch Cosmos Coalition for open-source world models

By

@runwayml

–

01 June 2026 7h13

Introducing the Cosmos Coalition A new global initiative with NVIDIA and leading AI labs to build and open-source frontier world models for physical AI. Runway joins as a founding member, working alongside NVIDIA and a set of leading AI labs to build, share and accelerate world

Cosmos 3 fully open: weights and post-training recipes available

By

–

01 June 2026 6h54

As always, Cosmos 3 is fully open. This includes model weights and post-training recipes. Available now on @huggingface

→ View original post on X — @simonw

April newsletter published late to show missed content

By

@simonw

–

01 June 2026 6h52

Each month, I publish the previous newsletter, now a month old, so people can see what they are missing. Here is what you would have received a month ago, covering what happened in LLMs in April: https://github.com/simonw/monthly-newsletter-archive/blob/main/2026-04-april.md
…

Nvidia AI generates F1 racing video from dashcam image

By

–

01 June 2026 6h52

Image-to-video generation is just as impressive.

Input image:
"Generate a 16:9 image from a dashcam view of a formula 1 racing event"

Video prompt:
"A high-speed racing event where a car navigates multiple winding turns"

🔊 Sound on – generated by Cosmos 3. pic.twitter.com/FxCWMVNDxh
— NVIDIA AI (@NVIDIAAI) 1 juin 2026

Image-to-video generation is just as impressive. Input image:
"Generate a 16:9 image from a dashcam view of a formula 1 racing event" prompt:
"A high-speed racing event where a car navigates multiple winding turns" Sound on – generated by Cosmos 3.

Cosmos 3 excels at multimodal reasoning, simulation, and robot training

By

–

01 June 2026 6h46

In addition to understanding and reasoning across modalities, Cosmos 3 excels at simulating physical environments, predicting future world states, and helping train robots to perform specific tasks.

It can do subsecond vision reasoning, large scale synthetic data generation and… pic.twitter.com/UJ8dIEG5ND
— NVIDIA AI (@NVIDIAAI) 1 juin 2026

In addition to understanding and reasoning across modalities, Cosmos 3 excels at simulating physical environments, predicting future world states, and helping train robots to perform specific tasks. It can do subsecond vision reasoning, large scale synthetic data generation and

NVIDIA AI ranks first on physical AI benchmarks among open models

By

–

01 June 2026 6h46

Delivering leading results on physical AI benchmarks among open models, it ranks first across @ArtificialAnlys
, Physics-IQ, PAI-Bench and R-Bench for world generation accuracy, RoboLab and RoboArena for action policy and the VANTAGE-Bench and TAR leaderboards for vision

Cosmos 3 unifies world generation, physics, and scene control

By

–

01 June 2026 6h46

Cosmos 3 ties everything together. Previous releases separated world generation, physical understanding, and controlled scene generation. Cosmos 3’s MoT architecture unifies these capabilities by pairing an autoregressive reasoner tower with a diffusion-based generator tower.

NVIDIA AI launches Cosmos 3, first fully open omnimodel for Physical AI

By

–

01 June 2026 6h46

Introducing Cosmos 3: Our latest frontier model for Physical AI

Cosmos 3 is the world’s first fully open omnimodel with native vision reasoning, world and action generation.

Today we’re releasing Super (32B) and Nano (8B) variants. pic.twitter.com/6UfkSA7kzQ
— NVIDIA AI (@NVIDIAAI) 1 juin 2026

Introducing Cosmos 3: Our latest frontier model for Physical AI Cosmos 3 is the world’s first fully open omnimodel with native vision reasoning, world and action generation. Today we’re releasing Super (32B) and Nano (8B) variants.