TencentARC/Looking-Ahead
Looking-Ahead WISA latent demos Pre-encoded example data for Looking-Ahead TF and FCP recipes: 100 distinct WISA video identities, each encoded for two Wan families, giving 200 latent records. These are loader-format examples, not the full historical training corpus or an evaluation benchmark. Source videos are not included. Contents Family Compatible recipes Source A Source B Disk [C,T,H,W] Loader [T,C,H,W] wan21_t2v_1.3b TF/FCP 1.3B 50 50 [16,22,54… See the full description on the dataset page: https://huggingface.co/datasets/TencentARC/Looking-Ahead.
| Author / organization | TencentARC |
|---|---|
| License (as declared) | apache-2.0 |
| Downloads | 0 (as reported by Hugging Face) |
| Likes | 1 |
| Created | 21 October 2026 |
| Last updated at source | 21 October 2026 |
Recent changes
- 21 Oct 2026 Description: none → Looking-Ahead WISA latent demos Pre-encoded example data for Looking-Ahead TF and FCP recipes: 100 distinct WISA video identities, each encoded for two Wan families, giving 200 latent records. These are loader-format examples, not the full
- 21 Oct 2026 Topics added task_categories:text-to-video, wisa, webdataset, latent removed video-generation, image-to-video
- 21 Oct 2026 Task: image-to-video → none
- 21 Oct 2026 Topics added video-generation, image-to-video removed task_categories:text-to-video, wisa, webdataset, latent
- 21 Oct 2026 Task: none → image-to-video
- 21 Oct 2026 Description: none → Looking-Ahead WISA latent demos Pre-encoded example data for Looking-Ahead TF and FCP recipes: 100 distinct WISA video identities, each encoded for two Wan families, giving 200 latent records. These are loader-format examples, not the full
- 21 Oct 2026 Topics added task_categories:text-to-video, wisa, webdataset, latent removed video-generation, image-to-video
- 21 Oct 2026 Task: image-to-video → none
- 21 Oct 2026 Topics added video-generation, image-to-video removed task_categories:text-to-video, wisa, webdataset, latent
- 21 Oct 2026 Task: none → image-to-video
- 21 Oct 2026 Description: none → Looking-Ahead WISA latent demos Pre-encoded example data for Looking-Ahead TF and FCP recipes: 100 distinct WISA video identities, each encoded for two Wan families, giving 200 latent records. These are loader-format examples, not the full
- 21 Oct 2026 Topics added task_categories:text-to-video, wisa, webdataset, latent removed video-generation, image-to-video
Changes detected by Silytics between refreshes of the source.
More from TencentARC
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 21 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.