Seedance 2.5

AI model · Video generation · Text → Video by Higgsfield

Advanced 4K AI video generation with native audio and multimodal control.

Seedance 2.5 is a high-resolution AI video generation model for creators and businesses, producing up to 30 seconds of 4K cinematic content with native audio from text or image prompts.

Model specs

The full spec card (context, pricing per token, modalities and more) on AntSpace. Sign up free to see the full spec card.

Price
From 0.11 USD 77 / s
Languages
English, Korean

Category: Generative AI

Modalities
Text, Image, Video, Audio → Video, Audio
Resolutions
Up to 4K resolution; 16:9, 9:16, 1:1 aspect ratios
Output formats
Video with optional native audio
Max output
30 seconds
Access
API, Higgsfield Creative Suite
Max Duration
30 seconds
  • Generates videos up to 30 seconds long
  • Supports 4K resolution output
  • Multimodal inputs: text, image, video, or audio
  • Optional native audio generation

Overview

Seedance 2.5 is Higgsfield's advanced AI video generation model designed for professional creators, marketing agencies, and filmmakers. It enables the creation of high-fidelity video clips up to 30 seconds long from a variety of inputs, including text prompts, reference images, and source videos. The model distinguishes itself by supporting 4K resolution output and the ability to generate native, synchronized audio alongside the visuals. Available through the Open Higgsfield API and integrated into the Higgsfield creative suite, it serves as a foundational tool for cinematic production and high-impact marketing content. Seedance 2.5 provides granular control over camera movement and framing, making it suitable for enterprise-scale creative workflows.

Multimodal Generation

  • Text to Video: Seedance 2.5 generates high-fidelity video clips from text descriptions, allowing users to define subjects, settings, and actions.
  • Image to Video: The model turns static photos, artwork, and product shots into moving scenes by using the image as a starting frame.
  • Video to Video: Users can provide a reference video to guide the motion, camera work, and timing of the generated output.
  • Audio-Enabled Generation: The model can optionally generate synchronized native audio and sound effects alongside the visual content.

Technical Capabilities

  • 30-Second Max Duration: Seedance 2.5 supports the creation of video clips with a maximum duration of 30 seconds.
  • 4K Resolution Output: The model produces high-resolution video output at up to 4K quality for professional filmmaking.
  • Flexible Aspect Ratios: Users can specify aspect ratios such as 16:9, 9:16, and 1:1 to fit different platform requirements.
  • API Integration: The model is available via a standard REST API for integration into third-party applications and creative workflows.

Use cases

  • Filmmakers generate cinematic scenes up to 30 seconds long at 4K resolution for professional productions.
  • Marketing teams turn product photos into high-fidelity video ads with synchronized native audio.
  • Content creators use video-to-video motion transfer to apply specific movements to new AI-generated characters.

Specifications

Modalities
Text, Image, Video, Audio → Video, Audio
Resolutions
Up to 4K resolution; 16:9, 9:16, 1:1 aspect ratios
Output formats
Video with optional native audio
Max output
30 seconds
Access
API, Higgsfield Creative Suite
Max Duration
30 seconds
Max Resolution
4K
Input Modalities
Text, Image, Video, Audio
Output Modalities
Video, Audio
API Price
$0.1177 per second
Aspect Ratios
16:9, 9:16, 1:1

How it works

  1. Get API Access: Developers can sign up for the Open Higgsfield API platform to receive an API key.
  2. Configure Generation: Send a POST request to the Seedance 2.5 endpoint with your prompt, resolution, and desired duration.
  3. Generate and Download: The model processes the multimodal inputs to generate a high-resolution video file with optional audio.

Pricing

API Usage
$0.1177 / s

FAQ

What is the maximum video length for Seedance 2.5?

Seedance 2.5 can generate video clips up to 30 seconds in length.

Can I use an image as a starting point for Seedance 2.5?

Yes, Seedance 2.5 supports generating videos from text, images, existing videos, or audio inputs.

What is the maximum resolution supported by Seedance 2.5?

Seedance 2.5 is capable of producing professional-grade video output at 4K resolution.

How much does it cost to use Seedance 2.5 via API?

Seedance 2.5 is priced at $0.1177 per second of generated video through the Higgsfield API.

Explore model

Added by AntSpace from public sources (open.higgsfield.ai).

Related

More AI models by Higgsfield