PixVerse/v5.5

PixVerse V5.5 was released on December 1, 2025. It is the first AI video large model to achieve "storyboard + audio" one-click direct output. This update supports the simultaneous generation of audio and multiple shots for the first time, and strengthens the ability to synchronize audio and video for multiple characters. AI can generate short films with narrative structures based on prompt words, marking the evolution of AI videos from "shot generation" to automatic "storytelling." Users can generate a 5-second, 8-second or 10-second video when multi-camera and audio-visual synchronization are turned on, and control the sound effect type directly in the prompt word. In terms of resolution, 10-second clips support up to 720p, and 1080p output is limited to 5 or 8 seconds. V5.5 adds new API parameters such as generate_audio_switch and generate_multi_clip_switch, providing developers with more fine-grained control capabilities.

Context
0 tokens
Max output
0 tokens
Input
llm image video
Output
video

Pricing

Price: $0.005 / request

360p without single : $0.225 /

360p without single : $0.45 /

360p without single : $0.495 /

540p without single : $0.225 /

540p without single : $0.45 /

540p without single : $0.495 /

720p without single : $0.3 /

720p without single : $0.6 /

720p without single : $0.66 /

1080p without single : $0.6 /

1080p without single : $1.2 /

360p with single : $0.275 /

360p with single : $0.5 /

360p with single : $0.545 /

540p with single : $0.275 /

540p with single : $0.5 /

540p with single : $0.545 /

720p with single : $0.35000000000000003 /

720p with single : $0.65 /

720p with single : $0.71 /

1080p with single : $0.65 /

1080p with single : $1.25 /

360p without multi : $0.375 /

360p without multi : $0.6 /

360p without multi : $0.645 /

540p without multi : $0.375 /

540p without multi : $0.6 /

540p without multi : $0.645 /

720p without multi : $0.45 /

720p without multi : $0.75 /

720p without multi : $0.81 /

1080p without multi : $0.75 /

1080p without multi : $1.35 /

360p with multi : $0.425 /

360p with multi : $0.65 /

360p with multi : $0.6950000000000001 /

540p with multi : $0.425 /

540p with multi : $0.65 /

540p with multi : $0.6950000000000001 /

720p with multi : $0.5 /

720p with multi : $0.8 /

720p with multi : $0.86 /

1080p with multi : $0.8 /

1080p with multi : $1.4000000000000001 /

Endpoints

  • openai

More models

  • dreamina-seedance-2-0-260128

    Dreamina Seedance 2.0 是字节跳动 Seed 团队开发的音视频生成模型,并集成到 CapCut 中。该模型采用双分支扩散 Transformer 架构,在生成链路中实现视觉与听觉信息流的并行处理,通过注意力/桥接机制实现帧级精度的声画同步。模型支持文本、图像、音频和视频多模态输入,最高可同步解析 12 个参考文件,生成长达 15 秒的视频,具备原生音视频同步生成能力。模型配备了隐形水印和 C2PA 内容凭证等保护措施。

  • PixVerse/v5.6

    PixVerse V5.6 于2026年1月26日发布,核心升级是“惊人的物理拟真”能力和电影级画面生成。模型在物理运动模拟方面实现显著进步,物体和角色的运动轨迹更符合真实世界的力学规律。在口型同步和音频真实感方面也进行了增强。V5.6支持文生视频和图生视频统一模型,分辨率覆盖360p至1080p,时长支持5秒、8秒和10秒。10秒视频仅适用于360p、540p和720p分辨率,1080p时长限制为5或8秒。模型还支持首尾帧过渡生成和提示词推理增强,在 Artificial Analysis 2026年2月的全球图生视频赛道中,V5.6以1291的 ELO 综合质量评分位列第四。

  • dreamina-seedance-2-0-fast-260128

    dreamina-seedance-2-0-fast 是 Seedance 2.0 的速度优化版本,继承了核心功能的同时大幅提升生成速度。它具备原生音频同步、摄像机控制和稳定动作能力,每次渲染成本更低,适合对生成速度有较高要求的电影剪辑场景。输出分辨率限制为 480P 和 720P(不支持 1080P)。在 Seedance 2.0 的产品矩阵中,Fast 版位于标准版和 Mini 版之间,是兼顾速度与质量的中档选择。

  • happyhorse-1.1-i2v