BaiLian

Wan3.0 Video

AlibabaToken-basedDynamic Pricing
Alias:wan3.0-video
Create API Key

Compare wan3.0-video API pricing, supported endpoints, capabilities and access options on Modelsell.

视频生成图生视频运镜控制多模态参考
Starting price
View full pricing
Context
—
Maximum input window
Modalities
→

Pricing by Supplier

Alibaba
-40%
阿里巴巴百炼官方
480p · Input + output duration
Per second$0.044118$0.026471/ second
720p · Input + output duration
Per second$0.088235$0.052941/ second
1080p · Input + output duration
Per second$0.176471$0.105882/ second

Capabilities / Supported modalities

Function callingToolsJSON modeStructured outputVision
Input
Output

Provider & data privacy

Provider
Unknown
Tokenizer
BPE (vendor-specific)
License
Provider-specificUnknown
Data retention4 daysNot used for upstream training by default

Performance

Benchmarks

Scores on standardized evaluations. Higher percentages are better — and rank percentile shows

Metrics sourced fromArtificial Analysis 2026-09-30·Wan3.0 Video

This model is not included in the current benchmark snapshot.

Missing models or measurements are not zero scores.

Benchmark charts preserve the source model selection and reasoning settings. Missing models or measurements are not zero scores, and benchmark cost or speed is not this site’s service commitment.

About Wan3.0 Video

一张满意的图片,往往已经决定了短片的主体、色彩和气氛。Wan 3.0(万相 3.0)可以从这个画面出发,继续补上动作和镜头:让商品在光线中缓缓转动,让人物回头看向镜头,或让一幅安静的风景出现云层、流水与景深变化。对已经有商品图、角色设定或分镜图的创作者来说,这是把静态素材变成视频的直接方式。

把想法写成一个镜头

图生视频不需要把原图重新描述一遍。更有用的是交代接下来发生什么、镜头如何移动,以及哪些细节要留下。万相 3.0 适合制作商品特写、人物动态、空间展示和氛围镜头;推近、环绕或轻微跟拍,都可以围绕同一张起始画面展开。动作描述越清楚,越容易判断生成结果是否符合你的分镜。

开场和收尾,都能提前安排

除了首帧,还可以提供尾帧引导镜头的结束状态。例如从商品侧面过渡到正面,或让角色从坐姿起身,最后停在你选好的构图。首尾帧的主体和场景最好能够自然衔接;它们帮助控制故事走向,但不等于逐帧锁定每一个动作。

为发布场景准备画面与声音

支持 2–30 秒视频、480p、720p 和 1080p 输出,也可以选择横屏、竖屏或方形画幅。需要完整氛围时,可以生成声音,并在提示词中交代风声、脚步或环境声;已有配乐和后期方案时,也可以只制作画面。商品详情页、社交媒体短片、概念预告和广告分镜,都可以从几秒钟的单镜头开始。

如果画面里有细小文字、复杂手部动作或多个快速运动的主体,建议先用短镜头试拍。清晰的原图、克制的动作和明确的视觉重点,通常比一次塞入许多变化更容易得到可用素材。

Use cases and prompting

商品广告:让一张香水图动起来

上传主体清晰、标签可辨认的香水瓶图片,先做一段 5 秒镜头:

保持香水瓶的瓶身、瓶盖和标签不变。镜头缓慢推近,窗边的柔和光斑从瓶身一侧掠过,玻璃边缘出现自然反光。背景保持干净,不增加物体。整个镜头连续、平稳,只有轻微的室内环境声。

想展示结尾构图时,再加入一张正面的尾帧。修改结果时,一次只调整一个重点,例如镜头速度或光线方向,便于看出哪项描述起了作用。

人物照片:做一个自然的小动作

人物保持原有服装与发型,先望向窗外,再慢慢转头看向镜头,露出轻微笑容。发梢被微风轻轻吹动,镜头固定,背景和光线保持连贯。

图片已经很详细,还要写哪些内容? 重点写动作、运镜和环境变化,并指出必须保留的商品特征或人物细节。

什么时候值得加尾帧? 当你对镜头最终停在哪里有明确要求时,例如商品转到正面,或人物走到指定位置。

标签或脸部出现变形怎么办? 换用更清晰的原图,减小转动幅度,缩短镜头,并减少遮挡与同时发生的动作;重要细节仍需要逐帧检查。

API access

API documentation

wan3.0-video

Use the linked documentation for model-specific request parameters and examples.

Authentication

All requests must include Authorization: Bearer <TOKEN> header. Anthropic-formatted endpoints accept the x-api-key header instead.

Generate tokens from the Tokens page; you can scope them to specific models, groups, IPs, and rate-limits.

Rate limits

SupplierRPMTPMRPD
AlibabaUnlimitedUnlimitedUnlimited

No restriction

Frequently asked questions about wan3.0-video

What is wan3.0-video?

Compare wan3.0-video API pricing, supported endpoints, capabilities and access options on Modelsell.

How do I call wan3.0-video?

Create an API key with access to wan3.0-video, then use the exact model ID and a supported endpoint from the API access section. Request fields depend on the selected endpoint.

How is wan3.0-video priced?

Pricing depends on the selected provider group and the model billing unit. The current input, output, request, or media prices are shown on this page before sign-up.

How should I evaluate wan3.0-video for my project?

Start with the use cases and prompting guidance on this page, then evaluate the model with representative inputs from your project.