wan-2.7-i2v-spicyWan 2.7 Spicy is an image-to-video model built on Alibaba's Wan 2.7 and fine-tuned for relaxed content moderation. Give it a first frame and a prompt and it returns a 2 to 15 second clip at 720p or 1080p with a generated stereo audio track — ambience, effects or voice — already inside the MP4. Pass your own audio file as audio_url and the model animates to it instead; set audio to false for a silent file. The latest upstream build made the audio fuller and the motion smoother, added anime-style and Asian-face scenes, and halved the prompt-rewriting step, so a one-line prompt is still expanded into camera, lighting and sound detail before generation (prompt_extend, on by default and free). Billing is per second of the duration you request: $0.169/s at 720p and $0.26/s at 1080p — a 5-second 1080p clip is $1.30, a 15-second one $3.90 — and failed requests are never charged. Measured on this platform: a 1080p, 2-second clip came back in 45 seconds as 1920x1080 at 30fps with AAC 44.1kHz stereo, no watermark. Where it fits: this is the premium tier of the Spicy line. Wan 2.2 Spicy is the volume engine — silent, 5 or 8 seconds, from $0.03/s, the right tool for game sprites, chat avatars, social loops and feed covers. Wan 2.7 Spicy is the clip you sell: pay-per-view and subscription content where the native audio does the immersion without a post step, branching interactive stories where every choice is one 2–15 second call, creator tools whose users publish straight from the API, and marketplaces where 1080p plus sound is the quality bar more affordable models cannot reach. Limits to design around: exactly one first frame (no last-frame interpolation — use wan-2.2-i2v-spicy for that), duration is a whole number from 2 to 15, resolution defaults to 1080p, and with prompt rewriting on the scene can drift away from the first frame within a couple of seconds — turn prompt_extend off when the exact scene must hold. We add no moderation layer of our own; the model's own safety systems still apply upstream. Complying with the law and your platform's rules, and using only subjects who have consented, is your responsibility.
JPG / PNG (no transparency) / WebP, 240–8000 px per side, max 20MB
WAV / MP3, 2–30 s, max 15MB. Leave empty and the model generates its own audio.
Generated video will appear here
Provide URLs and click Generate
Ambience, effects or voice generated with the frames as AAC stereo — or pass audio_url and the clip animates to your own sound
Whole seconds, billed on what you request: $0.169/s at 720p, $0.26/s at 1080p. A 5-second 1080p clip is $1.30
A Spicy fine-tune of Wan 2.7. We add no filter of our own; the model's own safety systems still apply upstream
Added in the latest upstream build, together with fuller audio, smoother motion and a prompt rewriter that runs in half the time
The premium tier of the Spicy line for pay-per-view clips, interactive stories and creator tools. Wan 2.2 Spicy is the silent volume engine from $0.03/s
Wan 2.7 Spicy is a Video Generation API provided by Alibaba. Wan 2.7 Spicy is an image-to-video model built on Alibaba's Wan 2.7 and fine-tuned for relaxed content moderation. Give it a first frame and a prompt and it returns a 2 to 15 second clip at 720p or 1080p with a generated stereo audio track — ambience, effects or voice — already inside the MP4. Pass your own audio file as audio_url and the model animates to it instead; set audio to false for a silent file. The latest upstream build made the audio fuller and the motion smoother, added anime-style and Asian-face scenes, and halved the prompt-rewriting step, so a one-line prompt is still expanded into camera, lighting and sound detail before generation (prompt_extend, on by default and free). Billing is per second of the duration you request: $0.169/s at 720p and $0.26/s at 1080p — a 5-second 1080p clip is $1.30, a 15-second one $3.90 — and failed requests are never charged. Measured on this platform: a 1080p, 2-second clip came back in 45 seconds as 1920x1080 at 30fps with AAC 44.1kHz stereo, no watermark. Where it fits: this is the premium tier of the Spicy line. Wan 2.2 Spicy is the volume engine — silent, 5 or 8 seconds, from $0.03/s, the right tool for game sprites, chat avatars, social loops and feed covers. Wan 2.7 Spicy is the clip you sell: pay-per-view and subscription content where the native audio does the immersion without a post step, branching interactive stories where every choice is one 2–15 second call, creator tools whose users publish straight from the API, and marketplaces where 1080p plus sound is the quality bar more affordable models cannot reach. Limits to design around: exactly one first frame (no last-frame interpolation — use wan-2.2-i2v-spicy for that), duration is a whole number from 2 to 15, resolution defaults to 1080p, and with prompt rewriting on the scene can drift away from the first frame within a couple of seconds — turn prompt_extend off when the exact scene must hold. We add no moderation layer of our own; the model's own safety systems still apply upstream. Complying with the law and your platform's rules, and using only subjects who have consented, is your responsibility. Through APIMODELS platform, you can access this model via a unified API with transparent pay-as-you-go pricing. Current pricing: 720p: $0.169, 1080p: $0.26.
Quickly generate brand promotion videos for ad campaigns and social media marketing.
Create compelling short-form video content for platforms like TikTok, Instagram, and YouTube.
Generate product feature demonstrations and tutorials to improve user conversion.
Produce course explanations, knowledge explainers, and training videos at an affordable price.
Wan 2.7 Spicy is available through APIMODELS at: 720p: $0.169, 1080p: $0.26. Billing is pay-as-you-go — you only pay for what you generate.
Sign up at APIMODELS, get your API key, and call our unified API endpoint. We provide detailed API documentation with code examples in cURL, Python, and Node.js.
APIMODELS offers the same Wan 2.7 Spicy model through our aggregation platform. We provide a unified API interface so you do not need separate accounts for each provider - one API key to access all models.
Per second of the duration you request: $0.169/s at 720p and $0.26/s at 1080p. A 5-second clip is $0.845 or $1.30, 10 seconds $1.69 or $2.60, 15 seconds $2.535 or $3.90. The audio track, a driving audio file, negative prompts and prompt rewriting cost nothing extra, and failed requests are not charged. Omitting resolution renders and bills at 1080p — pass 720p explicitly to pay less.
They complement each other. **Wan 2.2 Spicy** is the volume engine: $0.03/s at 480p and $0.06/s at 720p, 5 or 8 seconds only, **silent**, with first-to-last-frame interpolation — the right tool for breathing and blinking sprites in game dialogue, a library of mood clips behind a chat avatar, looping social shorts, animated feed covers, VTuber loop backgrounds and tip reactions. **Wan 2.7 Spicy** is the premium tier: 720p or 1080p, any whole number of seconds from 2 to 15, **native audio**, no last frame — for pay-per-view and subscription clips, interactive stories where each branch is one call, creator tools whose users publish straight from the API, and marketplaces where 1080p plus sound is the quality bar. Same endpoint, same key; only the model field changes.
Yes. By default the MP4 carries a stereo AAC track the model generates to match the scene (44.1kHz, 2 channels in our test). Pass audio_url — an https link to a 2–30 second WAV or MP3, at most 15MB — and the model drives the clip from your sound instead, with that audio in the output. Pass audio: false for a silent file, even if audio_url was given.
It is a community fine-tune of Alibaba's Wan 2.7 with looser content filtering than the official line; the latest upstream build also added anime-style and Asian-face scenes. We add no moderation layer of our own — prompts and reference images pass through as sent — but the model's own safety systems still apply upstream and screen both inputs and outputs, so this is not an unfiltered endpoint. Reviewing what you generate, complying with the law and your platform's rules, and using only subjects who have consented is your responsibility.
The first frame is a JPEG, PNG (no transparency), BMP or WebP with each side between 240 and 8000 pixels, an aspect ratio between 1:8 and 8:1 and at most 20MB, as a public URL or base64; exactly one frame — a last_image returns a 400. Duration is a whole number from 2 to 15; 5.5 or 16 returns a 400 rather than being rounded and billed. Measured here, a 1080p 2-second clip took 45 seconds; longer and sharper clips take longer. One more thing: with prompt rewriting on, the scene can drift away from the first frame within a couple of seconds — pass prompt_extend: false when the exact scene must hold.
On APIMODELS, Wan 2.7 Spicy runs alongside 60+ models on one API key and one balance, so choosing is about fit, not lock-in. It supports Image to Video, Native Audio, 2–15s, 720p / 1080p, Audio-driven, Relaxed Moderation, and you can weigh it on price and capability against other Video Generation models, then switch by changing a single model-name string — no new account or integration. Browse every Video Generation option with live pricing at apimodels.app/models.
Wan 2.7 Spicy supports: Image to Video, Native Audio, 2–15s, 720p / 1080p, Audio-driven, Relaxed Moderation. See the APIMODELS docs for full parameters and call examples.
Yes. APIMODELS exposes Wan 2.7 Spicy through a single unified API and one key — no separate provider accounts, and no need to handle each provider's regional network access yourself.
We support Stripe (Visa, Mastercard, and other international cards) and Alipay. Credits are available instantly after payment.
Prompts shared by their authors — copy and adapt them. Each one credits its author and links back to the original post.
Graduation-gown close-up: a worried monologue to camera
这是一个正面的近景镜头，画面主要呈现一位年轻黑人女孩的头部和上半身，她穿着一件蓝色的毕业袍。镜头稍微偏一点点，可以看到左边还有一个人，但背景被虚化处理了，看起来很模糊。女孩正对着镜头说话，嘴巴在动，眉头稍微皱着，表情看起来有点严肃和忧虑。光线从上方照下来，把她脸上的轮廓衬托得很自然，甚至能看到皮肤真实的质感。镜头虽然固定，但带着一点点人手持拍摄时的轻微呼吸感和晃动，视线从看着前面慢慢变成了低头看手里。
by Wan 3.0 视频创作者手册 · 通义万相
Suspense long take down an empty hospital corridor
一段现代悬疑风格的电影感长镜头，整个画面呈现出强烈的淡绿色滤镜风格，营造出一种基于淡绿色系的清冷氛围。在散发着冰冷白光的荧光灯下，空无一人的医院走廊呈现出低饱和的色彩与极高对比的光影关系，锐利且非人情化。镜头从护士台后方开始，一名身穿墨绿色夹克的女子猛地从走廊尽头的门后冲入，沿着对称的中心线向镜头快速跑来，摄影机流畅地向后移动。她的脚步声急促而响亮，口中焦急地呼喊。最终，她冲到台前，双手用力拍在台面上，绝望地环顾四周，发出几近崩溃的求助呐喊，声音在只有微弱设备嗡鸣的寂静中回荡。
by Wan 3.0 视频创作者手册 · 通义万相
One-take comedy: director versus leading lady on set
镜头1 [0秒-30秒] 近景，平视视角，连续快速摇镜，一镜到底（单镜头连续无缝运镜）： 画面左侧是身穿黑色潮流导演马甲、戴着高档专业工作耳麦的导演，其左下方放置着一台巨大的专业电影摄像机。右侧是妆容精致、留着时尚韩式逗号刘海、身穿高定制西装的关系户小鲜肉。背景是宽敞的专业影棚拍摄现场，可见巨大的绿色幕布、数个高耸的影视级C型灯架、缠绕的电线，以及在背景中穿梭忙碌的灯光师和化妆师等工作人员。 专业影棚人工光。柔和的LED柔光箱光源从右侧洒入，将三位主角的面部均匀照亮，场景色彩饱满。背景的摄影器材与绿幕在充足的影棚灯光下投射出淡淡的阴影，呈现出真实而忙碌的现代化大制作片场质感。 镜头手持拍摄一镜到底，近景，镜头一开始，拍摄导演，导演无所谓的表情，瞥了一眼画面右侧斜前方画外的一眼关系户小鲜肉，对他下达指令，说道：“媒体故意黑你，做个表情看看”。说完后镜头快速右摇到关系户面部近景，镜头不要切镜，关系户小鲜肉，面部朝向画面左侧斜前方画外导演方向，神色自若，眼神真挚而自信。他嘴唇微动，神情专业地向对方阐述到““就这种事情来说，情绪可以有好几种”，说话时，头部伴有极其轻微的自然晃动，说完后镜头快速左摇到导演面部近景，镜头不要切镜，导演思索了一下，一边挑衅似地歪了歪头，向关系户小鲜肉说到““狗仔爆料你深夜幽会”，导演说完台词后，镜头迅速右摇到小鲜肉面部近景，不要切镜，关系户小鲜肉，面部朝向画面左侧斜前方画外导演方向，并快速进入表演状态，他眉头紧锁，眼神慌乱、焦急地快速游移，随后他狠狠地咬紧下唇，闭上双眼，整张脸（虽带着精致妆容）因极度的焦虑而紧绷，完美呈现出面对偶像生涯毁灭时，内心惊恐、紧张交织的挣扎状态。表演完后， 镜头再次迅速左摇到导演近景，不要切镜，导演面部表情微变，语速加快，说到““媒体又帮你洗白了，说那是你亲姐！”导演说完后，镜头再次迅速右摇到小鲜肉近景，镜头不要切镜，镜头中，关系户小鲜肉，面部朝向画面左侧斜前方画外导演方向，他的表情在千分之一秒内完成神级转换。他先是面部一滞，随后双眼猛地眯起，嘴角极大程度地咧开，露出一排白皙的牙齿，展露出一个极其夸张、甚至有些变形的狂喜笑容。这种“星途尽毁的绝望”与“彻底摆脱束缚”交织的疯魔情绪在他脸上扭曲呈现。角色表演完后，镜头再次迅速左摇到导演近景，镜头不要切镜，导演说到“你亲姐向媒体透露，你是领养的”，导演说完后，镜头再次迅速右摇到小鲜肉近景。镜头不要切镜，镜头内，关系户小鲜肉面部朝向画面左侧斜前方画外导演方向，只见他双眼暴睁，眼球凸出，精心修剪的眉毛高高扬起，嘴巴震惊地张成“O”形。紧接着，他的嘴角再次无法克制地向上拉扯，露出一副不可思议、惊魂未定却的复杂神态。 角色表演完后，镜头再次快速左摇到导演近景，镜头不要切镜，紧接着导演露出有点惊讶的表情说到“其实你是千亿豪门真少爷”。导演说完，镜头再次随即迅速右摇到小鲜肉近景，镜头不要切镜，镜头内，关系户小鲜肉面部朝向画面左侧斜前方画外导演方向，他瞬间爆发出极度魔性的狂喜。他整个人激动得微微颤抖，双眼笑得眯成了一条缝，双手甚至抬起至胸前，双手小幅度快速连续鼓掌，身体前倾摇晃，将一夜暴富、无法自持的癫狂喜悦演绎得淋漓尽致。 角色表演完后，镜头再次快速左摇到导演近景，镜头不要切镜，导演说到“你姐卷款跑路”导演说完，镜头再次迅速右摇到小鲜肉近景。小鲜肉面部朝向画面左侧斜前方画外导演方向，只见他的双眼再次瞪大，瞳孔颤抖，嘴角向下耷拉，神情在一秒钟之内从天堂坠入地狱，写满了惊愕、痛苦与不知所措。画面最后定格在这个表情上。
by Wan 3.0 视频创作者手册 · 通义万相
We curate copy-ready prompt libraries — every entry shows its full text and a sample result, ready to adapt.