StartupRedwood City, CA, United States
Trains the Dyna series of robot foundation models (Dyna-1, the Dyna-2 world-action model) deployed on commercial dexterous tasks.
Dyna Robotics is a robotics company based in Redwood City, California, that builds general-purpose robots powered by its own embodied-AI foundation models (DYNA-1). It was co-founded by Lindon Gao and York Yang, who previously founded Caper AI, together with former Google DeepMind research scientist Jason Ma. It raised a $23.5M seed round in March 2025 and a $120M Series A in September 2025.
Sources: prnewswire.com (25 Mar 2025) · roboticstomorrow.com (16 Sep 2025)
Private company · $144M disclosed in USD across 2 of 2 listed rounds
| Date | Round | Amount | Lead | Source |
|---|---|---|---|---|
| 25 Mar 2025 | Seed | $23.5M | CRV, First Round Capital | prnewswire.com, 25 Mar 2025 |
| 15 Sep 2025 | Series A | $120M | Robostrategy, CRV, First Round Capital | therobotreport.com, 16 Sep 2025 |
#71 of 104 model companies · score 13.8 / 100 · updated 19 Oct 2026
Each bar shows an input from 0 to 100% before weighting. 0 of 3 listed models have open weights.
The score (0–100) measures public, checkable evidence about a company's robot models. It does not measure private capability. A company that publishes no benchmark results and no weights can still have strong models. Every input is computed from the data on each company page, and every number there links to its source.
| Input | Weight | How it is measured |
|---|---|---|
| Published benchmarks | 35% | For each scored benchmark, the company's best result divided by the best result any company has published, averaged over the benchmarks it reports. No published results scores 0. Results come only from the model's own paper, model card or the official benchmark leaderboard. Scored: LIBERO (4-suite average success, %); SimplerEnv Google Robot, visual matching (%); SimplerEnv Google Robot, variant aggregation (%); SimplerEnv WidowX / Bridge (%); CALVIN ABC→D (average length, 0–5); RoboArena (pairwise rating). |
| Open weights | 20% | Share of the company's listed models whose weights the public can download. |
| Adoption | 20% | GitHub stars on the company's most-starred official model repository, on a log scale relative to the most-starred repository in the list. Hugging Face downloads are shown on company pages but not scored, because gated repositories report zero. |
| Released models | 15% | Number of publicly announced, named models, capped at 5. |
| Disclosed funding | 10% | The sum of the funding rounds listed on the company's profile that were disclosed in USD, on a log scale from $10M to the largest sum in the list. Rounds reported only in other currencies are listed but not converted or counted. This input applies only to private startups and robot makers. For big-tech divisions, research labs and listed companies it is n/a, and its weight is shared across the other inputs in proportion. |
Sources. arXiv papers and official model cards for benchmarks and parameters; official leaderboards (RoboArena); the GitHub and Hugging Face APIs for stars and downloads; company announcements and established press for funding. Missing values are shown as n/a, never estimated.
Updated 19 Oct 2026. Benchmark protocols differ between papers, for example in fine-tuning and number of trials, so small gaps between close scores are not meaningful.
Closed weightsPolicy
Three-layer system (RL whole-body controller, Dyna-2 policy, vision-language workflow orchestrator) for hour-long mobile manipulation workflows on the Taku semi-humanoid.
No benchmark results on a shared public benchmark have been published for this model.
Closed weightsWorld model
World-action model on a video-diffusion backbone that jointly denoises future video and actions, pretrained on up to 1 million hours of data.
No benchmark results on a shared public benchmark have been published for this model.
Closed weightsPolicy
Foundation model for continuous dexterous tasks such as napkin and laundry folding in commercial deployments; a pre-trained base version was described in Nov 2025.
No benchmark results on a shared public benchmark have been published for this model.
