Seedance 2.0 R2V

• 🏢 ByteDance's flagship reference-to-video model in the Seedance 2.0 line.
• 🆕 Built on Seed's unified multimodal architecture accepting text, images, video, audio.
• 📏 Generates coherent, audio-synchronized multi-shot clips in a single pass.
• 🎯 Combines multiple reference files spanning images, videos, and audio.
• 👁️ Native audio-video joint generation with cinematic camera and lighting control.
• 🔧 Supports video editing and extension while preserving motion and style.
• ⚡ Released March 2026; Fast and Enhanced variants also exist in the family.
• 💬 Can directly reference text-based storyboards and shooting scripts.

ByteDance

ByteDance is a Chinese internet technology company headquartered in Beijing, widely known as the parent company behind TikTok and Douyin. Beyond social media, ByteDance has invested heavily in artificial intelligence research, building generative media models…

Seedance 2.0 R2V is the reference-to-video endpoint of ByteDance's flagship Seedance 2.0 video family, released in March 2026. ByteDance describes Seedance 2.0 as built on Seed's unified multimodal architecture, which accepts text, images, video clips, and audio as inputs and produces coherent, audio-synchronized video in a single pass. The R2V mode is the most flexible Seedance 2.0 configuration, letting creators combine multiple reference files across modalities and weave them into one cinematic output with native audio, real-world physics, and director-level camera control.

Beyond generation from scratch, the model handles targeted editing and video extension — modifying specified clips, characters, or actions while preserving original motion and camera work, or continuing a scene with consistent characters and style. ByteDance notes it can even reference text-based storyboards and shooting scripts to direct visual composition, camera language, and motion rhythm.

Compared with its predecessor Seedance 1.5 Pro, ByteDance describes Seedance 2.0 as a substantial upgrade, citing improved instruction-following, more stable subject consistency, and better motion and physics for complex character interactions, plus the addition of dual-channel audio-video output. Technical figures here are vendor-reported; consult ByteDance's official Seedance 2.0 documentation for precise parameters.

Within the family, R2V sits alongside the text-to-video Seedance 2.0, the speed-oriented Seedance 2.0 Fast R2V, and the higher-fidelity Seedance 2.0 Enhanced R2V, part of the broader Seed lineup that also includes the Seedream V5 Lite image models.