Superfast Video
Superfast Video is a proprietary, closed-weight text-to-video model developed by LINEARGAME PTE. LTD. and made available under the YorollAI brand. It generates 1344×768, 24 fps videos with audio through a hosted API.
API-only availability: model weights, training code, inference code, and downloadable checkpoints are not distributed through this repository.
Try the model
The Hugging Face Space is a hosted evaluation interface. It sends requests to the official Superfast API; inference does not run from model files stored in this repository.
Capabilities
| Property | Current hosted preview |
|---|---|
| Task | Text-to-video |
| Output resolution | 1344×768 |
| Frame rate | 24 fps |
| Audio | Included |
| Durations | 5, 10, or 15 seconds |
| Controls | Text prompt and seed |
| Access | Hosted API only |
Intended use
Superfast Video is intended for creative video generation, concept visualization, prototyping, marketing creatives, social video, entertainment, and other lawful uses permitted by the applicable service plan and the Yoroll Terms of Use.
Commercial use
- The public Hugging Face Space is provided for evaluation, demonstration, and non-commercial testing.
- General commercial use is available to eligible paid API customers, subject to the applicable service plan and the Yoroll Terms of Use.
- High-volume use, white-label or resale use, regulated or high-risk use cases, and uses requiring custom service levels or contractual assurances require a separate enterprise agreement.
- No access or license to the model weights, inference implementation, or other proprietary model technology is granted.
Subject to applicable law and the governing service terms, YorollAI does not claim ownership of a user's eligible output solely because it was generated using the service. Copyrightability varies by jurisdiction and by the degree of human contribution. Outputs are not guaranteed to be unique, copyrightable, or free of third-party rights.
Restrictions and responsible use
Users must have all rights and permissions required for their prompts and other inputs. The service must not be used for illegal content, fraud, deceptive impersonation, privacy or publicity-rights violations, infringement of intellectual property, sexual exploitation, or other uses prohibited by the Yoroll Terms of Use or applicable law.
Users are responsible for reviewing generated content before publication or commercial deployment. Extra review and safeguards are necessary when content depicts real people, brands, protected characters, sensitive events, or other third-party rights.
Limitations
- Generated video may contain visual, temporal, physical, textual, or audio inconsistencies.
- The service does not guarantee factual accuracy, identity consistency, uniqueness, fitness for a particular purpose, or non-infringement.
- Generation is asynchronous and may include queueing time.
- Result links returned by the current hosted preview are temporary.
- Capabilities, limits, and output characteristics may change as the hosted service is updated.
Model and training information
The architecture, parameter count, model weights, training code, detailed training data composition, and production inference implementation are proprietary and are not published in this repository. Evaluation results will be added when a public, reproducible evaluation methodology is available.
License and terms
Superfast Video is closed-weight and proprietary. This repository is an informational model card and does not distribute or license model artifacts. Access to the hosted service and use of generated outputs are governed by the Yoroll Terms of Use, the applicable service plan, and any separately executed agreement.