Seedance 2.0 test shows strong Genshin-style video generation with character refs
aziz4ai · x · 2026-07-28
The post says Seedance 2.0 can generate impressive outputs when prompted with “in the style of Genshin Impact.” The author tested both text-to-video and character-reference workflows, including 4K-to-720p results, and says the reference images still influence the output while the overall quality remains gorgeous. The tip is to avoid prompt words that may trigger other styles.
Related event: Seedance 2.0 Tested: Shows Impressive Video Generation Capabilities(2 posts)→
More from Multimodal
- NVIDIA’s Sol-Attn speeds up video generation inference by 2.1x — Haopeng Li · 2026-07-28
- Users ask how to keep characters consistent in LTX2.3 image-to-video workflows — Adventurous-Gold6413 · 2026-07-28
- A game character walking a dog turns into a cute multimodal mashup — Humble-Tangerine-878 · 2026-07-28
- A repeatable AI anime workflow uses character sheets, storyboards, GPT Image 2.0 and Seedance 2.0 — aftahi_ai · 2026-07-28
- Spanish-Language AI Media Generation Video Tutorial Series Launched, from Basics to Advanced — Botoni · 2026-07-28
- Leaked Imagine Omni upgrade would bring images, video and voice into Grok Imagine — XFreeze · 2026-07-28