Seedance 2.0 Mini R2V

• 👁️ Reference-to-video endpoint of ByteDance's lightweight Seedance 2.0 Mini tier.
• 🔧 Accepts images and video clips as combined multimodal reference input.
• 🎯 Built for character identity and scene consistency across multi-shot output.
• 💬 Family generates joint audio-video, including dual-channel audio.
• 📏 Seedance 2.0 supports up to 15-second multi-shot generations.
• ⚡ Mini sits alongside the full Seedance 2.0 and Fast variants.
• 🆕 Adds editing and video-extension controls introduced with Seedance 2.0.
• 🏢 Released June 2026; newer Seedance 2.5 R2V now exists alongside it.

ByteDance

ByteDance is a Chinese internet technology company headquartered in Beijing's Haidian district, with an associated variable-interest entity incorporated in the Cayman Islands. Best known as a consumer internet operator at enormous scale, it has also become a…

Seedance 2.0 Mini R2V is the reference-to-video ("R2V") endpoint of ByteDance's lightweight Seedance 2.0 Mini tier, released in June 2026. Where text-to-video starts from a prompt alone, R2V lets you attach reference material — images and video clips — and point to it inside the prompt, so the model can carry over a character's face, an outfit, a camera move or a motion rhythm.

The capability set comes from the Seedance 2.0 generation. ByteDance describes 2.0 as supporting combined multimodal reference input, understanding elements such as visual composition, camera language and motion rhythm, and even referencing text-based storyboards directly. Compared with the earlier Seedance line — represented here by Seedance 1.5 Pro — ByteDance reports a substantial upgrade in controllability, with better instruction following and more stable subject consistency for complex, interaction-heavy scenes, plus new targeted editing of clips, characters and storylines and continuous-shot video extension. The generation also adds 15-second multi-shot audio-video output with dual-channel audio.

Within the 2.0 line, Mini is the lightweight tier rather than the full-fidelity one. It sits beside Seedance 2.0 R2V and the speed-focused Seedance 2.0 Fast R2V, and shares its tier with Seedance 2.0 Mini text-to-video and Seedance 2.0 Mini image-to-video. For the newest reference-to-video generation, see Seedance 2.5 R2V.