BindWeave: Subject-Consistent Video Generation
BindWeave is a unified subject-consistent video generation framework by ByteDance Research. Built on MLLM-DiT architecture for single and multi-subject prompts (ICLR 2026).
Guides and workflows related to this model series.
No articles found.