Flora-captioner-27B
Flora-captioner-27B generates one English action instruction for a target interval in a robot or human manipulation video, using neighboring video context to disambiguate the operation. NVIDIA describes it as a research and development model for non-commercial use, released to help researchers scale up custom robot datasets.
What it does
Researchers and developers can use it to annotate manipulation-video intervals with action instructions and to prototype robotics data-curation pipelines.
- Availability
- Hugging Face release date: 10/15/2026
- Pricing model
- Not established
- Deployment
- Global; Linux; PyTorch with Hugging Face Transformers
- Privacy
- Not established
- Licence
- NVIDIA OneWay Noncommercial License (22 March 2022)
Capability
Modality
Access
Deployment
Pricing
Target user
Evidence
6 claims- Generates one action instruction for a supplied target interval in a robot or human manipulation video. [1]
Flora-captioner-27B generates one action instruction for a supplied target interval in a robot or human manipulation video.
NVIDIA. Quoted text is outside CC BY 4.0 and keeps its original rights. - Uses neighboring video context to disambiguate the intended local operation. [1]
It uses neighboring video context to disambiguate the intended local operation.
NVIDIA. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model is released on Hugging Face. [1]
Flora was developed by NVIDIA as the data engine used for training GR00T-2, and is released to help researchers scale up custom robot datasets.
NVIDIA. Quoted text is outside CC BY 4.0 and keeps its original rights. - The release date listed for Hugging Face is 10/15/2026. [1]
- The model supports Linux and runs with PyTorch and Hugging Face Transformers. [1]
- The model is intended for non-commercial robotics research and development and is not intended or supported for commercial or production use. [1]
Flora is intended for non-commercial robotics research and development only, including annotating robot and human demonstration videos.
NVIDIA. Quoted text is outside CC BY 4.0 and keeps its original rights.
Sources
1
1
Verified
Versions
2| Version | Effective | Change | Changed fields | State |
|---|---|---|---|---|
| 2 | 2026-10-23 04:43 UTC | material change | modality, availability, deployment | Current |
| 1 | 2026-10-22 05:32 UTC | publication | None | Current |
In briefs
2- A toolkit for voice agents that listen, respond, and take action Weekly brief 2026-W43
- A model writes action labels for robot videos Daily brief 2026-10-22
Original catalog content is licensed under CC BY 4.0, attributed to wowsuchai. Source quotations and linked content keep their original rights.