Wan 2.2 Plus overview
Wan 2.2 Plus by Alibaba Cloud
Wan 2.2 T2I Plus API provides high-quality text-to-image generation for creative and AI-powered visual applications.
Wan 2.2 Plus specs
- Model ID
alibaba/wan2.2-t2i-plus- Developer
- Alibaba Cloud
- Input
- Text
- Output
- Image
- Endpoint
POST https://api.aimlapi.com/v1/images/generations
Wan 2.2 Plus API Overview
Wan 2.2 Plus builds upon previous versions by significantly improving image stability during generation, ensuring fewer artifacts and more consistent outputs. It excels at producing richly detailed visuals with artistic flair suitable for a broad spectrum of creative applications.
Technical Specifications
- Model Type: Transformer-based Text-to-Image Generator
- Input Modalities: Text prompts (natural language)
- Output Modalities: High-resolution images (up to 1024x1024 pixels)
- Architecture: Advanced multi-layer transformer with integrated attention mechanisms focused on fine-grained visual features
- Training Data: Diverse large-scale multimodal datasets combining artistic, natural, and synthetic images
- Parameter Count: Approximately 2.5 billion parameters
- Inference Speed: Optimized for real-time and batch processing on high-end GPUs
- Supported Languages: Primarily English with robust handling of descriptive natural language
Performance Benchmarks
- FID Score: Achieves a Fréchet Inception Distance (FID) of 12.3, indicating high image realism and quality (lower is better)
- CLIP Score: Average CLIP embedding similarity score above 0.31, reflecting precise semantic alignment between input text and generated image
- Generation Consistency: Less than 5% variation in repeated generations under fixed random seed conditions
- Speed: Generates 512x512 images in under 1.2 seconds on a single NVIDIA A100 GPU
Key Features
- Enhanced Stability: Reduced image artifacts, consistent rendering quality across multiple generations on the same prompt
- High Detail Resolution: Superior handling of complex textures, fine lines, and subtle visual elements
- Artistic Style Adaptation: Flexible output style ranging from photorealism to abstract artistic effects
- Multimodal Robustness: Effective interpretation of nuanced text, including adjectives and complex scene descriptions
- Cross-Domain Creativity: Capable of combining multiple concepts and styles within a single coherent image
- Scalable Output Size: Supports variable output resolutions suited for web, print, or digital art
Comparison with Other Models
vs Stable Diffusion V2: Wan 2.2 Plus delivers higher image stability and fewer visual artifacts compared to Stable Diffusion V2. While Stable Diffusion offers more flexibility in prompt styles and wider community support, Wan 2.2 Plus excels in rendering fine details and natural expressions consistently.
vs DALL·E 3: DALL·E 3 is renowned for its strong abstract concept generation and creative variety, but Wan 2.2 Plus provides sharper control over photorealistic detail and artistic style blending. Wan 2.2 Plus produces consistently realistic images with enhanced stability, making it better suited for nuanced realism.
vs Midjourney V5: Midjourney V5 favors surreal and ethereal art styles with high creativity, whereas Wan 2.2 Plus focuses on finely detailed, naturalistic images with fewer artifacts. Wan 2.2 Plus has an edge in preserving texture details and subtle visual elements, making it ideal for realistic imagery.
Use cases
- Digital Art Creation: Artists can generate original illustrations, concept art, and visual storytelling elements with minimal input.
- Content Marketing: Quickly produce eye-catching visuals for social media, blogs, or advertisements tailored to text themes.
- Design Prototyping: Aid designers in visualizing products, environments, or characters during early development stages.
- Creative Entertainment: Generate backgrounds, character inspirations, or scenes for games, films, and animations.
- Educational Tools: Support creative teaching methods by illustrating complex concepts or historical scenes dynamically.
Wan 2.2 Plus API pricing
| Usage | Price | Unit |
|---|---|---|
| Image generation | $0.05 | per generation |
Prices from the AI/ML API price list, checked on Nov 4, 2026.
Wan 2.2 Plus vs other image models: prices
AI/ML API rates for 5 models, per generation.
| Model | Per generation |
|---|---|
 Wan 2.2 Plus | $0.05 |
| Qwen Image 2.0 | $0.035 |
| Qwen Image 3 | $0.075 |
| Qwen Image 3 Edit | $0.075 |
| Qwen Image 2.0 Pro | $0.075 |
How to use the Wan 2.2 Plus API
View docs
Build with your agent
Give your agent the model details, code and documentation.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v1/images/generations' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "alibaba/wan2.2-t2i-plus",
"prompt": "A ginger cat asleep on a stack of exactly three hardcover books in a rainy bookshop window at dusk. Gold lettering on the glass reads OPEN LATE. Warm lamp light inside, raindrops and blurred neon reflections outside, 35mm film photo."
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "alibaba/wan2.2-t2i-plus",
"prompt": "A ginger cat asleep on a stack of exactly three hardcover books in a rainy bookshop window at dusk. Gold lettering on the glass reads OPEN LATE. Warm lamp light inside, raindrops and blurred neon reflections outside, 35mm film photo."
}
response = requests.post("https://api.aimlapi.com/v1/images/generations", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v1/images/generations", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "alibaba/wan2.2-t2i-plus",
"prompt": "A ginger cat asleep on a stack of exactly three hardcover books in a rainy bookshop window at dusk. Gold lettering on the glass reads OPEN LATE. Warm lamp light inside, raindrops and blurred neon reflections outside, 35mm film photo."
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());Models similar to Wan 2.2 Plus
Related articles
.webp)
Wan 2.2 Plus FAQ
What inputs and outputs does Wan 2.2 Plus support?
Wan 2.2 Plus takes image, text as input and returns image.
When was Wan 2.2 Plus released?
Wan 2.2 Plus became available on November 13, 2025.
How much does Wan 2.2 Plus cost?
Wan 2.2 Plus costs $0.05 per generation on AI/ML API, billed as you go with no subscription.
How is Wan 2.2 Plus billed?
Wan 2.2 Plus is billed per generation, a fixed charge per output rather than by prompt length.
Can Wan 2.2 Plus take an existing image as input?
Yes, Wan 2.2 Plus accepts image input as well as a text prompt.
Who created Wan 2.2 Plus?
Wan 2.2 Plus was built by Alibaba Cloud.
How do I call Wan 2.2 Plus on AI/ML API?
Send a request to https://api.aimlapi.com/v1/images/generations with alibaba/wan2.2-t2i-plus as the model id.
Is Wan 2.2 Plus available through a single API key?
Yes. Wan 2.2 Plus is served through AI/ML API, so the same key and endpoint format used for other models applies.
What improvements does Wan 2.2 Plus offer over the previous Wan 2.1 model?
Wan 2.2 Plus offers improved stability and a higher success rate compared to Wan 2.1.
What is the Wan 2.2 Plus model ID?
Use alibaba/wan2.2-t2i-plus as the model parameter. Requests go to https://api.aimlapi.com/v1/images/generations with your AI/ML API key.