Nvidia Cosmos
Shares rank 4 of 5 world models by recommenders
Physics-aware simulation and video generation models that people find realistic and useful for training data. Main caveat: they are too large to run on most consumer computers.
Physics-aware simulation and video generation models that people find realistic and useful for training data. Main caveat: they are too large to run on most consumer computers.
8 people gave an opinion
- recommend it
- 5
- criticize it
- 3
- approval share
- 63%
Recommended in 2 of 13 months
- November 2025: not recommended
- December 2025: not recommended
- January 2026: not recommended
- February 2026: not recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: not recommended
- June 2026: recommended
- July 2026: not recommended
- August 2026: not recommended
- September 2026: not recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 5 people recommend itNeeds 10 or more, not met.
- 1.6 recommenders for every criticNeeds 2 to 1 or better, not met.
- Recommended in 2 of 13 monthsNeeds 4 or more, not met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Realistic physics and video output
“simulation and video generation with realistic physics is kind of changing the game”
Comment from 19 Oct 2026 in “How to Train Your Robot” “These demos honestly look pretty good to me.”
Comment from 1 Jun 2026 in “Nvidia Cosmos 3”
What people warn about
Each mark is one comment
Too big for consumer hardware
“too big to run on most people's computers”
Comment from 1 Jun 2026 in “Nvidia Cosmos 3” “Looking forward to trying this out on my $10000+ workstation grade GPU that I need an equally expensive set up to run.”
Comment from 1 Jun 2026 in “Nvidia Cosmos 3” Doubts about approach and examples
“I doubt that there's a lot of juice to squeeze with this kind of approach IMO.”
Comment from 1 Jun 2026 in “Nvidia Cosmos 3” “Most of the examples they've chosen seem.. not good?”
Comment from 1 Jun 2026 in “Nvidia Cosmos 3”
Other world models people recommend
The four most recommended other world models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Marble by World Labs | 25 | 10 | 71% | 6 | — | Approved |
| Project Genie | 16 | 17 | 48% | 1 | — | — |
| World Labs world model | 7 | 5 | 58% | 4 | — | — |
| Nvidia CosmosThis page | 5 | 3 | 63% | 2 | — | — |
| LingBot-World | 5 | 0 | 100% | 3 | — | — |
What people said
All 8 comments that mention it, newest first. Each links to the original comment.
“simulation and video generation with realistic physics is kind of changing the game”
“Cosmos 2 in particular already has taken the image diffusion world by storm”
“These demos honestly look pretty good to me.”
“They produce exactly the kind of in-domain data that already exists in droves.”
“I doubt that there's a lot of juice to squeeze with this kind of approach IMO.”
“Most of the examples they've chosen seem.. not good?”
“Looking forward to trying this out on my $10000+ workstation grade GPU that I need an equally expensive set up to run.”
“too big to run on most people's computers”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.