Wan 2.7 Text-to-Video Spicy API

Wan 2.7 Text-to-Video Spicy is based on Alibaba Cloud Tongyi Wanxiang's Wan 2.7 multimodal video generation model, in its Spicy mode (uncensored improved version) with relaxed content restrictions compared to the standard version.

Spicy is a channel naming created by the iCreat platform. The model capability comes from the original Wan 2.7 base model, and Spicy is the platform's own name for its uncensored variant. It is NOT an official Alibaba release.

On the iCreat platform, the model is served through the endpoint https://api.icreat.ai/v1/task/submit/aliyun/wan2-7/text-to-video-sp, using a two-step asynchronous flow: submit a task, then poll for the result. It offers 720P and 1080P resolution tiers and is billed per output second at $0.10/sec for 720P and $0.15/sec for 1080P.

Model Positioning

Wan 2.7 Text-to-Video Spicy is a text-to-video generation model positioned for creators who need uncensored, prompt-driven video synthesis without supplying any reference image. It targets the same production pipeline as the standard Wan 2.7 Text-to-Video model, but with relaxed content restrictions, making it suitable for creative, artistic and experimental work that would otherwise be constrained.

Because it is a Spicy channel on iCreat, the model should be understood as a platform-provided variant rather than an official Alibaba product line. The underlying generation capability, resolution tiers, aspect ratios, duration range and billing model are identical to the standard Wan 2.7 Text-to-Video model; only the content restriction level differs.

Core Capabilities

Text-to-Video Generation Without Reference Images

Wan 2.7 Text-to-Video Spicy generates video from a text prompt alone. No reference image or first frame is required. The model supports both Chinese and English prompts and can directly describe subject, action, camera movement, lighting atmosphere and visual style.

Smart Prompt Expansion

Long Chinese and English prompts are natively supported, and the model can intelligently expand a brief instruction by filling in scene details. This lowers the barrier to writing effective prompts.

Negative Prompt Control

The negative_prompt field lets you specify content that should not appear in the generated video. This is useful for suppressing unwanted objects, artifacts, styles or motions, and gives finer control over the final output without rewriting the main prompt.

Native Audio-Visual Coordination and Lighting Rendering

Wan 2.7 delivers native audio-visual coordination and sound-effect sync, together with strong motion continuity, realistic physics simulation and high-quality lighting rendering. The result is video whose motion connects naturally and whose texture approaches live-action footage, suited to commercial advertising, short dramas and animation production where output quality matters.

Multiple Resolutions and Aspect Ratios

The model supports 720P and 1080P resolution tiers together with five aspect ratios: 1:1, 16:9, 9:16, 4:3 and 3:4. This covers square, landscape and portrait delivery formats, so a single model can serve social, web and broadcast-oriented compositions.

Adjustable Duration

Output duration is adjustable from 4 to 15 seconds in integer steps. The default is 5 seconds. This allows short clips for social media as well as longer single-shot sequences while keeping billing predictable per output second.

Asynchronous Two-Step Task API

The model is exposed as an asynchronous task API. Submitting a request returns a task_id, which is then used to poll the task result endpoint. Successful results include a costUSD field so that the cost of each generation can be tracked programmatically.

Spicy Uncensored Variant

While preserving the generation capability of the base model, the content restrictions are relaxed, providing a usable channel for creative directions such as horror, suspense and fringe subject matter that are hard to pass under standard moderation.

Pricing

Total Cost = Unit Price × Output Video Duration.

Note: Official international pricing matches the platform (720P $0.10/sec, 1080P $0.15/sec). Where differences exist, iCreat platform pricing prevails. Prices listed here reflect iCreat AI's standard rates. Some models may be eligible for promotional discounts or special offers. Final charges are based on the real-time prices displayed in each model's Playground.

Application Scenarios

• Short-form social video generated purely from a written prompt
• Creative and artistic video experiments that require relaxed content restrictions
• Marketing and product concept clips produced without shooting or stock footage
• Storyboard and previsualization for film, advertising and game cinematics
• Multilingual content production using Chinese or English prompts
• Rapid iteration on camera movement, lighting and style ideas
• Automated video pipelines that submit tasks and poll results programmatically

Model Comparison

Note: The Spicy and standard versions share the same underlying model, resolution tiers, aspect ratios, duration range and billing. The only difference is the level of content restriction.

Note: All three generations bill per output video second, but they differ in resolution tiers, duration control and price structure. Wan 2.7 Spicy supports continuous integer-second duration control, Wan 2.6 Spicy uses fixed duration tiers, and Wan 3.0 Spicy adds a low-cost 480P tier with longer single-run durations. Prices are subject to real-time platform rates.

Why Choose Wan 2.7 Text-to-Video Spicy?

• Uncensored Spicy channel with relaxed content restrictions for creative freedom
• Prompt-only text-to-video generation, no reference image required
• Smart prompt expansion supports both Chinese and English input
• Five aspect ratios and two resolution tiers cover most delivery formats
• Flexible 4–15 second duration with integer-step control
• Native audio-visual coordination and sound-effect sync
• Simple asynchronous two-step API with task_id and costUSD tracking
• Tra