Wan 2.7 (Text→Video)
预览Alibaba Cloud全部 Alibaba Cloud 模型
$0.10
$0.15
2026-04-03
定位:均衡。价格:视频模型中第 2 级(共 4 级)。
- 定位
- 均衡。原厂没有给定位,按视频模型的价格带归类。 原厂已推出新一代。
- 价格
- 在视频模型中是第 2 级(共 4 级,$ 最便宜),看的是720p 每秒价格(没有 720p 就看最低一档):$0.09 到 $0.12 以下。
- 原厂速度
- 不标:原厂没有比较这款的速度。
三项各自独立,不是质量评分。
Wan2.7-t2v 是阿里巴巴 Wan2.7-Video 套件中的文生视频模型,于 2026 年 4 月与图生视频、参考生成及视频编辑等同系列模型一同推出。阿里巴巴 API 文档列出 2 到 15 秒的输出,镜头结构可在 prompt 中指定,包括标注时间码的多镜头清单。阿里巴巴也强调 FPV 无人机俯冲、360 度环绕等运镜。
5 秒 720p 约
$0.50
USD
720p
$0.10
USD / 秒
1080p
$0.15
USD / 秒
发布
2026 年 4 月 3 日
价格
用多少付多少:请求成功才扣点,失败不收费。
720p
$0.10
1080p
$0.15
USD / 秒
720p
$0.10
10 点数 · USD / 秒
1080p
$0.15
15 点数 · USD / 秒
按视频 Token 计价 · USD / 每百万
| 分辨率 | 文生视频 | 图/视频生视频 |
|---|---|---|
| 480p–720p | $4.6296 | $4.6296 |
| 1080p | $3.0864 | $3.0864 |
1080P · 5 s ≈ $0.75
文生视频(Wan 2.7)。 按输出分辨率每秒计费(video tokens:宽 × 高 × 秒数 × 24 fps ÷ 1024);参考与编辑模式的输入视频秒数一并计费。比照原厂牌价直通——无 ATP markup。 注意:此模型目前不论请求分辨率均输出 1080P,并按 1080P 计费。 任务失败不计费。Preview:正式开放前费率可能调整。
示范
全部是模型直接生成的原始结果,没有后期处理。展开任意一条就能看到完整提示词。
提示词
A large sheet of soft sand-coloured linen settles slowly over an unseen rounded form, folds forming and rippling outward as it comes to rest. Seamless warm-grey studio backdrop. Medium shot, single movement: slow dolly push-in as the fabric settles, eye level, 50mm lens, shallow depth of field. Raking side light from the left to carve the folds into relief, warm neutral grade, subtle 35mm film grain, 24fps film look. Negative: blurry, warped text, oversaturated, watermark.
提示词
A tidy retail shelf lined with unbranded matte cylindrical containers in graduated earth tones, all labels blank, price rails empty. Bright modern store interior. Wide shot, single movement: slow pan left-to-right along the shelf, eye level, 35mm lens, deep focus keeping the full row sharp. Even overhead retail lighting with soft fill, clean neutral grade, 24fps film look. Negative: blurry, distorted faces, warped text, readable labels, oversaturated, watermark.
提示词
A Korean man in his late 30s, short black hair swept back, clean-shaven, composed purposeful expression, tall lean build, wearing a fitted dark grey suit without a tie, walks to centre stage and turns to face the audience, a clicker in one hand. Behind him a vast screen shows only a soft neutral gradient. Silhouetted audience heads in the foreground. Wide shot, single movement: slow dolly push-in as he turns, eye level, 35mm lens, shallow depth of field. Stage keylight from the front with cool spill from the screen behind, high-contrast cinematic corporate grade, 24fps film look. Negative: blurry, distorted faces, warped text, readable slides, oversaturated, watermark.
提示词
A high-ceilinged distribution warehouse: plain cardboard cartons travel along a conveyor while a worker in a high-visibility vest scans one with a handheld reader, tall racking receding behind. Steady working energy. Wide shot, single movement: tracking shot at shoulder height following the cartons along the conveyor, 35mm lens, deep focus. Bright even industrial ceiling lighting with cool daylight from clerestory windows, neutral desaturated grade, 24fps film look. Negative: blurry, distorted faces, warped text, readable labels, oversaturated, watermark.
提示词
Shot 1: wide establishing view of a bright industrial assembly line, a single matte-gray articulated robotic arm mounted at its station, precision components moving along a conveyor below it. Shot 2: cut to medium close-up as the arm's gripper closes around a small metal bracket with a soft mechanical click, rotates it ninety degrees, and sets it into place on the moving line, motion steady and mechanically precise throughout. Clean white factory environment, evenly diffused overhead lighting, no visible people, no branding on equipment. Deep focus throughout, eye level, 35mm lens. Bright even industrial lighting, neutral clean grade, no film grain.
提示词
A generic clear glass perfume bottle with a matte black cap resting on a dark reflective turntable, the liquid inside a warm amber tone. It rotates slowly and steadily through a full turn, its facets catching a single rim light in a smooth sweep as it turns, the reflection below mirroring the motion. Vertical composition, medium close-up, locked-off static camera with only the turntable moving, eye level, macro lens, shallow depth of field with the black backdrop staying soft. Single hard rim light plus soft ambient fill, high-contrast studio grade, black seamless background.
适合场景
适合
- 在 prompt 中以时间码镜头清单撰写的多镜头分镜
- 仅凭文本生成 FPV 俯冲、环绕等电影感运镜
- 2 到 15 秒的社交或广告短片
通常较不适合
- 严格控制在 720P 预算——此 id 在 ATP 目前一律输出并按 1080P 计费
- 让已有图片动起来——ATP 上对应的是 Wan 2.7 图生视频(wan-2-7-i2v)
每个模型都用同一套管理:项目密钥、各项目可用的模型、点数预算,每次请求都有记录。设置组织
成本试算
拉一下用量,看一个月大概花多少。
分辨率档
07,200
试算
$30.00
USD / 月
3,000 点数 / 月
- 720P$30.00
$30.00 USD / 月 · 3,000 点数 / 月
快速上手
先确认你的密钥能看到这个模型代号,再按文档发请求。 阅读文档
models.sh · curl
curl -s https://api.atptoken.ai/v1/models \
-H "Authorization: Bearer atp-•••"
# data[].id: "wan-2-7-t2v"常见问题
- Wan 2.7 (Text→Video) 可以输入什么、会生成什么?
- Wan 2.7 (Text→Video) 是 Alibaba Cloud 的文生视频模型。输入文本提示,生成视频(720P / 1080P)。
- Wan 2.7 (Text→Video) 怎么计价?
- 按视频秒数计价,每秒价格取决于你选的分辨率(720P / 1080P)。价格区也列出各分辨率每百万视频 Token 的价格。请求成功才扣点,失败不收费。USD 1=100 点。 查看价格
- ATP 上有比 Wan 2.7 (Text→Video) 更新的 Alibaba Cloud 模型吗?
- 有。Wan 3.0 (Video) (wan-3-0-video) 于 2026 年 8 月 6 日 发布,比 Wan 2.7 (Text→Video)(2026 年 4 月 3 日)新,是 ATP 上较新的 Alibaba Cloud 视频模型。两个都能用:先在本页比较同家族价格,再到项目里开启想用的代号。
- 怎么通过 ATP 调用 Wan 2.7 (Text→Video)?
- 创建项目 API 密钥,先用 GET https://api.atptoken.ai/v1/models 确认能看到「wan-2-7-t2v」,再按视频生成文档发请求,model 填这个代号。
资料来源
本页的模型介绍引用以下来源。价格来自 ATP。