Seedance 2.0

• 🎯 ByteDance's flagship text-to-video model, built on unified multimodal architecture
• 🆕 Single-pass cinematic video with native audio synthesis
• 👁️ Accepts text, images, video clips, and audio as inputs
• 🔧 Director-level control over camera, lighting, and shadow
• 🎬 Real-world physics simulation for more realistic motion
• 📚 Emphasis on scene consistency and cinematic quality
• 🏢 Developed by ByteDance's Seed research division

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 is the flagship entry in ByteDance's Seedance text-to-video family, released in early 2026 and developed by the company's Seed research division. Built on a unified multimodal architecture, it accepts text, images, video clips, and audio as inputs and produces coherent, audio-synchronized video in a single pass. ByteDance positions it for cinematic output with native audio, real-world physics, and director-level control over camera, lighting, and shadow.

As the next-generation flagship, Seedance 2.0 succeeds the earlier same-family Seedance 1.5 Pro and the 1.0 line. ByteDance frames realistic motion, scene consistency, and cinematic quality as defining goals of the 2.0 release. The provider page presents it for filmmaking-style generation, with reference-driven workflows that condition output on supplied images, clips, and audio.

The model targets controllable, story-driven video rather than single short shots, combining motion fidelity with the unified handling of multiple input modalities described above. Specific clip lengths, resolutions, and per-request input limits vary by deployment and are documented by individual API hosts rather than confirmed here.

The family includes several derivatives optimized for different tradeoffs: Seedance 2.0 Fast for quicker generation, Seedance 2.0 Mini, and Seedance 2.0 Enhanced, plus image-to-video and reference-to-video variants like Seedance 2.0 and Seedance 2.0 R2V.