Veda-Sparse/Minimax-H3-T2VA-Veda-8NFE-600Step-Preview
archivedmodel
Veda is a learned sparse-attention method for video diffusion models. Dense. License: minimax-h3-community. Vendor-reported: "Visual and audio quality on par with full attention while skipping 90% of it."
- Vendor-reported: Visual and audio quality on par with full attention while skipping 90% of it.
- Org
- Veda-Sparse
- Released
- 2026-09-25
- Status
- archived
- License
- minimax-h3-community (restricted)
- Architecture
- dense
- Task
- text-to-video · video
- Signal
- 115 HF likes · 27665 downloads
- Archived
- 2026-10-25 (aged out (older than 30 days); stats as last seen)
Links
First published here 2026-09-25 · updated 2026-10-25 · markdown · JSON