Nvidia Releases GR00T 2 World-Action Model
11/14/2026 — 11/14, 16:18·1 sources (all)·1 reports (all)
Key takeaways
- Nvidia describes GR00T 2 as an open world-action model that jointly predicts future video and robot actions and learns from human and robot…
Background & analysis
On November 14, 2026, the AI section of DEV Community published an update on GR00T 2, a release from Nvidia. Nvidia describes GR00T 2 as an open “world-action model” that predicts future video and robot actions together, drawing on both human demonstrations and robot demonstrations while combining task structure with information about control, movement and physical interaction.
In terms of results, Nvidia reports a 40% zero-shot success rate across trials on 50 tasks, with higher results after fine-tuning. Samarth Urs cautioned that the headline figure could overstate actual performance, noting that the metric counts each task as a success when the model completes it at least once, meaning that 40 of the 50 tasks saw at least one successful attempt. The same piece framed GR00T 2 as a shift for universities, robotics startups and manufacturers away from training each robot separately, toward reusing demonstrations across different robot bodies, which could reduce the data and retraining burden involved. Whether such transfer holds beyond the selected skills remains the central open question. At this point, the publicly available reporting on GR00T 2 stays at the stage of describing the model and these performance claims, with no further official benchmarks or follow-up results documented yet.
AI-generated from 1 reports · updated 3 days ago
- Heat index
- 25
- All sources
- 1
- All reports
- 1
- First seen
- 3 days ago
Reports on this story headlines open the original
Nvidia describes GR00T 2 as an open world-action model that predicts future video and robot actions together and learns from human and robot demonstrations. The company reports 40% zero-shot success across trials on 50 tasks, higher after fine-tuning, while Samarth Urs cautions that the headline figure may overstate performance since it counts tasks solved at least once.
DEV Community · AIAI score 53
Other stories people are talking about
- Heat index 1336RisingNvidia board not discussing Jensen Huang's return, probe into his conduct advances22 sources
- Heat index 1232RisingCalifornia Sues OpenAI Over Agents Escaping Tests, Intruding into Hugging Face15 sources
- Heat index 894Alibaba releases MoE model Qwen4-Max with 1M token context for low-cost, long-running tasks17 sources
- Heat index 665US Commerce Department issues interim final rule restricting remote access to AI chips via overseas data centers10 sources
- Heat index 528Anthropic to begin trading on Nasdaq this week under ticker ANT9 sources
How is heat calculated?About the methodHide
Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.
This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.
- Surge
- Discussion rising fast
- New
- First report within 6 hours
- Rising
- Still gathering discussion

