muapi Logomuapi Logo
muapi Logomuapi Logo

Seedance 2 Spicy Omni Reference: 图像转视频

Generate bold multi-reference video with Seedance 2 Spicy Omni Reference. VIP priority routing, up to 4K, reduced content filtering. Try free. Seedance 2 Spicy Omni Reference by ByteDance. Same VIP-tier priority routing as Seedance 2 VIP Omni Reference, generating videos from up to 9 image references, 3 video clips, and 3 audio references, with reduced content-safety filtering for more creative freedom. Reference materials in your prompt with @image1…@image9, @video1…@video3, and @audio1…@audio3. Selectable output resolution (720p/1080p/4K) — price scales with the resolution chosen.

Interactive model controls

Seedance 2 Spicy Omni Reference by ByteDance. Same VIP-tier priority routing as Seedance 2 VIP Omni Reference, generating videos from up to 9 image references, 3 video clips, and 3 audio references, with reduced content-safety filtering for more creative freedom. Reference materials in your prompt with @image1…@image9, @video1…@video3, and @audio1…@audio3. Selectable output resolution (720p/1080p/4K) — price scales with the resolution chosen.

📝

概览

关于此模型

Seedance 2 Spicy Omni Reference is the relaxed-moderation sibling of Seedance 2 VIP Omni Reference. It generates video guided by up to 9 image references, 3 video clips, and 3 audio references, with the same VIP-tier priority routing and native audio-visual sync — but with reduced content-safety filtering for bolder, more dramatic creative freedom. Reference your uploaded materials in the prompt using @image1-@image9, @video1-@video3, and @audio1-@audio3 syntax. Output resolution is selectable directly on this endpoint — 720p, 1080p, or 4K — via the resolution parameter, and price scales with your choice. Supports character sheets via @character:<id> and trained characters via @omni-character:<char_id>.

1Bold Character Consistency: Maintain a character's appearance across multiple video scenes using reference images and @omni-character trained IDs, with a bolder visual treatment than the standard tier.
2Mature Creative Direction: Explore darker or more intense visual themes in AI-generated video art with multi-reference guidance.
3Audio-Synced Video: Generate video synchronized to reference audio clips for music videos, voiceover content, or rhythmic motion.
4Multi-Modal Reference: Combine images, video clips, and audio references in a single request for richly guided, high-contrast video generation.
5Style Transfer: Use reference images as visual style guides to generate new video scenes with a bolder, higher-contrast artistic style.
💰

价格与价值

成本分析

muapiapp$0.30-$1.35/秒 (720p-4K)

Same VIP-tier priority routing and multi-reference support as Seedance 2 VIP Omni Reference, with a bolder visual style; price scales with the selected resolution.

Fal.ai暂不可用

Seedance 2 Spicy 档位不可用。

Replicate暂不可用

Seedance 2 Spicy 档位不可用。

** 竞品价格根据相似模型架构和使用层级估算。

⚙️

技术详情

配置参数

提示词string

Video description. Use @image1…@image9 to 参考图像, @video1…@video3 for videos, and @audio1…@audio3 for audio. Use @character:<request_id> for a Seedance 2 character sheet or @omni-character:<char_id> for a trained omni-character. Multiple characters are supported.

默认值@image1 is the main character. The person walks along a city street at sunset, cinematic lighting.
分辨率枚举（3 个选项）

Output resolution. Price scales with resolution: 720p is cheapest, 4K is most expensive.

默认值720p
图像 URLarray

最多 9 个参考图像 URL（JPEG/PNG/WebP）。第 N 张图像对应提示词中的 @imageN。

默认值https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpg
参考视频 URLarray

最多 3 个参考视频片段 URL（MP4，每个最长 15 秒）。第 N 个视频对应提示词中的 @videoN。

默认值undefined
参考音频 URLarray

最多 3 参考音频 files (MP3/WAV, total max 15s). Each Nth audio corresponds to @audioN in the prompt.

默认值undefined
画面比例枚举（6 个选项）

输出视频的画面比例。

默认值16:9
时长（秒）int

视频时长（秒）。

默认值5
高比特率boolean

启用高比特率模式以获得更好的画面保真度，但会生成更大的文件。

默认值false
📖

实施指南

开发者文档

How to Use Seedance 2 Spicy Omni Reference

  1. Provide reference media: Supply up to 9 images, 3 video clips, and 3 audio files, referenced in your prompt via @image1...@image9, @video1...@video3, @audio1...@audio3.

  2. Choose resolution: Set resolution to 720p (default), 1080p, or 4k — price scales with resolution, identical to the VIP tier it's built on.

  3. Write your prompt and set duration: Describe the scene, and choose a duration between 4 and 15 seconds. Reduced filtering means fewer prompts get rejected, but baseline platform policy still applies.

  4. Submit and poll: The API returns a request_id. Poll /predictions/{request_id}/result or use a webhook URL.

curl -X POST https://api.muapi.ai/api/v1/seedance-2-spicy-omni-reference \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"@image1 walks along a city street at sunset, dramatic lighting","images_list":["https://example.com/a.jpg"],"resolution":"1080p","duration":5}'
❓

常见问题

常见问答

What makes Spicy different from Seedance 2 VIP Omni Reference?

Spicy uses the same priority routing, resolution options, and duration limits as VIP Omni Reference, but leans into a bolder, more dramatic visual style with reduced content-safety filtering. Pricing and output quality are identical to the underlying tier.

What is the @ reference syntax?

Reference uploaded images with @image1-@image9, video clips with @video1-@video3, and audio with @audio1-@audio3, in the order they appear in each corresponding list.

What resolutions does Seedance 2 Spicy Omni Reference support?

Set `resolution` to `720p` (default), `1080p`, or `4k`. Price scales with resolution, at the same rate as Seedance 2 VIP Omni Reference.

Can I still get my request blocked?

Reduced filtering means fewer false positives, but requests that violate baseline platform policy are still rejected.

What is the difference between @character and @omni-character?

@character:<request_id> references a character sheet from a completed Seedance 2 Character generation. @omni-character:<character_id> references a previously trained Omni Reference character. Both can be combined in the same prompt.