Bytedance family
# Seedance 2 Fast High-speed multimodal video generation with audio synchronization. ## Overview Seedance 2 Fast is a high-performance video generation model designed for marketing teams and creative professionals who require rapid iteration without sacrificing visual quality. By leveraging advanced multimodal architecture, this model excels at transforming text prompts, reference images, and audio inputs into cohesive, high-fidelity video content. It is specifically optimized for speed, making it an ideal choice for social media campaigns, rapid prototyping, and real-time content production workflows. What sets Seedance 2 Fast apart is its native integration of audio-driven animation and character consistency. Unlike standard video models that require post-production syncing, this model generates synchronized audio—including dialogue, sound effects, and background music—directly alongside the visual output. With support for multiple reference inputs, users can maintain strict brand identity and character consistency across various scenes, ensuring that the final output aligns perfectly with existing creative assets. ## Use Cases * **Social Media Ad Campaigns:** Generate high-quality, 5-second product showcases or promotional clips with synchronized music in seconds. * **Character-Driven Storytelling:** Maintain consistent character appearance across multiple video segments using reference images for brand-safe storytelling. * **Automated Lip-Sync Content:** Create engaging talking-head videos for influencer marketing or corporate communications by pairing reference audio with portrait images. * **Motion Transfer for Brand Assets:** Apply the motion style of a reference video to a static product image to create dynamic, eye-catching advertisements. * **Rapid Prototyping:** Quickly iterate on visual concepts for ad storyboards using the 'fast' inference mode to test multiple creative directions. ## Parameters Seedance 2 Fast offers a robust suite of controls to fine-tune your video output. Please refer to the input parameters section in the Pixloop interface to configure your seed, resolution, aspect ratio, and reference media inputs to achieve your desired creative outcome. ## Tips for Best Results * **Leverage Reference Inputs:** For maximum character consistency, upload up to 9 reference images and reference them explicitly in your prompt using the [Image1], [Image2] syntax. * **Use Intelligent Duration:** Set the duration to -1 to allow the model to automatically determine the optimal length for your prompt, which often results in more natural pacing. * **Audio-Driven Precision:** When using reference audio, ensure your prompt includes specific cues like [Audio1] to help the model align the visual motion with the provided sound file. * **Resolution Strategy:** Use 480p for rapid iteration and testing, then switch to 720p for final high-quality renders to optimize your workflow speed. * **Prompting for Dialogue:** If you need specific speech, use double quotes in your prompt to signal to the model that it should prioritize lip-syncing the text to the generated audio. ## About Seedance 2 Fast is provided by Bytedance and is a high-speed, optimized variant of the Seedance 2.0 architecture. It is designed to bridge the gap between complex multimodal generation and the high-velocity requirements of modern digital advertising. By utilizing the latest advancements in diffusion-based video synthesis, it offers a powerful toolset for creators looking to integrate AI-driven video production into their professional creative pipelines.
Model input reference derived from preset schema.
| Name | Type | Required | Default | Description |
|---|---|---|---|---|
| image | image | No | null | First frame image. Cannot be combined with reference images. |
| prompt | string | Yes | null | Text prompt for video generation |
| duration | integer | No | 5 | Video duration in seconds. Seedance 2.0 range [4,15], or -1 for intelligent duration. |
| resolution | select | No | "720p" | Video resolution. Seedance 2.0 Fast supports 480p/720p only. |
| aspect_ratio | select | No | "adaptive" | Video aspect ratio (maps to `ratio`). Default adaptive. |
| generate_audio | boolean | No | true | Generate synchronized audio. Put dialogue in double quotes in the prompt. |
| last_frame_image | image | No | null | Last frame image. Only works with a first frame image. Cannot be combined with reference images. |
| reference_audios | audio | No | [] | Up to 3 reference audios (total <= 15s). Requires >=1 reference image/video. Reference as [Audio1] etc. |
| reference_images | image | No | [] | Up to 9 reference images. Reference as [Image1] etc. No real human faces. |
| reference_videos | video | No | [] | Up to 3 reference videos (total <= 15s). Reference as [Video1] etc. |