
Wan 2.5 Image Edit API by Alibaba
Open and Advanced Large-Scale Image Generative Models.
程式碼範例
import requests
import time
# Step 1: Start image generation
generate_url = "https://api.atlascloud.ai/api/v1/model/generateImage"
headers = {
"Content-Type": "application/json",
"Authorization": "Bearer $ATLASCLOUD_API_KEY"
}
data = {
"model": "alibaba/wan-2.5/image-edit",
"prompt": "A beautiful landscape with mountains and lake",
"width": 512,
"height": 512,
"steps": 20,
"guidance_scale": 7.5,
}
generate_response = requests.post(generate_url, headers=headers, json=data)
generate_result = generate_response.json()
prediction_id = generate_result["data"]["id"]
# Step 2: Poll for result
poll_url = f"https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}"
def check_status():
while True:
response = requests.get(poll_url, headers={"Authorization": "Bearer $ATLASCLOUD_API_KEY"})
result = response.json()
if result["data"]["status"] == "completed":
print("Generated image:", result["data"]["outputs"][0])
return result["data"]["outputs"][0]
elif result["data"]["status"] == "failed":
raise Exception(result["data"]["error"] or "Generation failed")
else:
# Still processing, wait 2 seconds
time.sleep(2)
image_url = check_status()安裝
為您的程式語言安裝所需的套件。
pip install requests驗證
所有 API 請求都需要透過 API 金鑰進行驗證。您可以從 Atlas Cloud 儀表板取得 API 金鑰。
export ATLASCLOUD_API_KEY="your-api-key-here"HTTP 標頭
import os
API_KEY = os.environ.get("ATLASCLOUD_API_KEY")
headers = {
"Content-Type": "application/json",
"Authorization": f"Bearer {API_KEY}"
}切勿在客戶端程式碼或公開儲存庫中暴露您的 API 金鑰。請改用環境變數或後端代理。
提交請求
import requests
url = "https://api.atlascloud.ai/api/v1/model/generateImage"
headers = {
"Content-Type": "application/json",
"Authorization": "Bearer $ATLASCLOUD_API_KEY"
}
data = {
"model": "your-model",
"prompt": "A beautiful landscape"
}
response = requests.post(url, headers=headers, json=data)
print(response.json())提交請求
提交非同步生成請求。API 會傳回一個預測 ID,您可以用它來檢查狀態並取得結果。
/api/v1/model/generateImage請求主體
import requests
url = "https://api.atlascloud.ai/api/v1/model/generateImage"
headers = {
"Content-Type": "application/json",
"Authorization": "Bearer $ATLASCLOUD_API_KEY"
}
data = {
"model": "alibaba/wan-2.5/image-edit",
"input": {
"prompt": "A beautiful landscape with mountains and lake"
}
}
response = requests.post(url, headers=headers, json=data)
result = response.json()
print(f"Prediction ID: {result['id']}")
print(f"Status: {result['status']}")回應
{
"id": "pred_abc123",
"status": "processing",
"model": "model-name",
"created_at": "2025-01-01T00:00:00Z"
}檢查狀態
輪詢預測端點以檢查請求的當前狀態。
/api/v1/model/prediction/{prediction_id}輪詢範例
import requests
import time
prediction_id = "pred_abc123"
url = f"https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}"
headers = { "Authorization": "Bearer $ATLASCLOUD_API_KEY" }
while True:
response = requests.get(url, headers=headers)
result = response.json()
status = result["data"]["status"]
print(f"Status: {status}")
if status in ["completed", "succeeded"]:
output_url = result["data"]["outputs"][0]
print(f"Output URL: {output_url}")
break
elif status == "failed":
print(f"Error: {result['data'].get('error', 'Unknown')}")
break
time.sleep(3)狀態值
processing請求仍在處理中。completed生成完成。輸出已可取得。succeeded生成成功。輸出已可取得。failed生成失敗。請檢查錯誤欄位。完成回應
{
"data": {
"id": "pred_abc123",
"status": "completed",
"outputs": [
"https://storage.atlascloud.ai/outputs/result.png"
],
"metrics": {
"predict_time": 8.3
},
"created_at": "2025-01-01T00:00:00Z",
"completed_at": "2025-01-01T00:00:10Z"
}
}上傳檔案
上傳檔案至 Atlas Cloud 儲存空間並取得 URL,可用於您的 API 請求。使用 multipart/form-data 上傳。
/api/v1/model/uploadMedia上傳範例
import requests
url = "https://api.atlascloud.ai/api/v1/model/uploadMedia"
headers = { "Authorization": "Bearer $ATLASCLOUD_API_KEY" }
with open("image.png", "rb") as f:
files = {"file": ("image.png", f, "image/png")}
response = requests.post(url, headers=headers, files=files)
result = response.json()
download_url = result["data"]["download_url"]
print(f"File URL: {download_url}")回應
{
"data": {
"download_url": "https://storage.atlascloud.ai/uploads/abc123/image.png",
"file_name": "image.png",
"content_type": "image/png",
"size": 1024000
}
}輸入 Schema
以下參數可在請求主體中使用。
無可用參數。
範例請求主體
{
"model": "alibaba/wan-2.5/image-edit"
}輸出 Schema
API 傳回包含生成輸出 URL 的預測回應。
範例回應
{
"id": "pred_abc123",
"status": "completed",
"model": "model-name",
"outputs": [
"https://storage.atlascloud.ai/outputs/result.png"
],
"metrics": {
"predict_time": 8.3
},
"created_at": "2025-01-01T00:00:00Z",
"completed_at": "2025-01-01T00:00:10Z"
}Atlas Cloud Skills
Atlas Cloud Skills 將 300 多個 AI 模型直接整合至您的 AI 程式碼助手。一鍵安裝,即可使用自然語言生成圖片、影片,以及與 LLM 對話。
支援的客戶端
安裝
npx skills add AtlasCloudAI/atlas-cloud-skills設定 API 金鑰
從 Atlas Cloud 儀表板取得 API 金鑰,並設為環境變數。
export ATLASCLOUD_API_KEY="your-api-key-here"功能
安裝完成後,您可以在 AI 助手中使用自然語言存取所有 Atlas Cloud 模型。
MCP Server
Atlas Cloud MCP Server 透過 Model Context Protocol 將您的 IDE 與 300 多個 AI 模型連接。支援任何 MCP 相容的客戶端。
支援的客戶端
安裝
npx -y atlascloud-mcp設定
將以下設定新增至您 IDE 的 MCP 設定檔中。
{
"mcpServers": {
"atlascloud": {
"command": "npx",
"args": [
"-y",
"atlascloud-mcp"
],
"env": {
"ATLASCLOUD_API_KEY": "your-api-key-here"
}
}
}
}可用工具
API Schema
Schema 不可用Seedance 1.5 Pro
原生音視頻同步生成聲畫合一,一鏡到底
ByteDance 開創性的 AI 模型,通過單一統一流程同步生成完美同步的音頻和視頻。支持 8 種以上語言,實現毫秒級精準唇音同步的真正原生音視頻生成體驗。
Why Choose Wan 2.5?
More Affordable
Despite Google's recent price cuts, Veo 3 remains expensive overall. Wan 2.5 is lightweight and cost-effective, providing creators with more options while significantly reducing production costs.
One-Step Generation, End-to-End Sync
With Wan 2.5, no separate voice recording or manual lip alignment is needed. Just provide a clear, structured prompt to generate complete videos with audio/voiceover and lip sync in one go - faster and simpler.
Multilingual Friendly
When prompts are in Chinese, Wan 2.5 reliably generates A/V synchronized videos. In contrast, Veo 3 often displays "unknown language" for Chinese prompts.
Precise Character Recreation
Wan 2.5 excels at character trait restoration, accurately presenting character appearance, expressions, and movement styles, making generated video characters more recognizable and personalized for enhanced storytelling and immersion.
Artistic Style Rendering
Supports Studio Ghibli-style rendering, creating hand-painted watercolor textures and animation effects. Brings warm, dreamy visual experiences that enhance artistic appeal and storytelling depth.
Who Can Benefit?
Marketing Teams
Whether it's product launches, promotional campaigns, or brand marketing, Wan 2.5 helps you quickly generate high-quality videos, making creation easy and efficient.
- Product demos and tutorials without coordination headaches
- Social media marketing with multilingual subtitles and lip sync
- AI-generated content lets teams focus on strategy and creativity
Global Enterprises
Provides ideal content localization solutions for multinational companies, making creation easier and more efficient.
- Multilingual video support with prompt recognition
- One-click generation of lip-synced subtitles and voiceovers
- Fast content localization for global markets
Story Creators / YouTubers
Creators can leverage Wan 2.5 to improve video production efficiency while ensuring high-quality output.
- Immersive storytelling with precise character actions and expressions
- Higher publishing efficiency with reduced editing and post-production time
- Diverse content from short videos to animated story segments
Corporate Training Teams
Wan 2.5 makes corporate training more efficient and engaging.
- Professional videos replace boring text documents
- Quick creation of operational demos and training tutorials
- Consistent style and standardized output for global rollout
Creative Freelancers / Small Studios
Wan 2.5 lets creativity flow without expensive equipment or actors - AI generates everything efficiently.
- Experiment with diverse works from short films to social media content
- From inspiration to completion with "one-click generation"
- High-quality content without expensive equipment or professional actors
Educational Institutions / Online Course Creators
Transform creativity into reality without high costs - Wan 2.5 makes quality content production easy and economical.
- Experiment with various styles from short films to promotional videos
- Higher production efficiency from concept to finished product
- Quality content without expensive equipment or professional talent
核心功能
One-Step A/V Generation
Generate complete videos with synchronized audio, voiceover, and lip-sync in a single process
Dual Character Sync
Supports simultaneous generation of two characters with synchronized actions, expressions, and lip-sync for natural interactions
Professional Quality
High-quality video output with realistic character expressions and precise lip synchronization
Multilingual Support
Excellent support for Chinese prompts and reliable generation of multilingual content
Cost Effective
Significantly lower costs compared to competitors while maintaining professional quality
Character Trait Restoration
Precisely recreates character appearance, expressions, and movement styles with high fidelity and personality
Artistic Style Rendering
Supports various artistic styles including Studio Ghibli-inspired hand-painted watercolor textures
Immersive Scenes
Perfect for dialogue scenes, interviews, or dual-person short films with natural audio-visual consistency
Wan 2.5 Prompt Showcase
Discover the power of Wan 2.5 through these curated examples. From digital human lip-sync to dual character scenes, artistic rendering to character restoration - experience the possibilities.
Study Room Scholar
Middle-aged man reading with perfect lip-sync in a warm study environmentA middle-aged man sitting at a wooden desk in a cozy study room, surrounded by bookshelves and a warm lamp glow. He opens an old book and reads aloud with a calm, deep voice: 'History teaches us more than just facts… it shows us who we are.' The room has subtle background sounds: pages turning, the faint ticking of a clock, and distant rain against the window.
Park Sunset Romance
Couple interaction with synchronized dual character actions and expressionsA young couple sitting on a park bench during sunset. The woman leans her head on the man's shoulder. He whispers softly: 'No matter where we go, I'll always be here with you.' The sound includes the rustling of leaves, distant laughter of children playing, and the gentle hum of cicadas in the evening air.
Ballet Performance Art
Precise character trait restoration with artistic movement and expressionA graceful ballerina with her hair in a messy bun, performing a powerful and emotional contemporary ballet routine. She is in a minimalist, dark art studio. Abstract patterns of light and shadow, projected from a hidden source, dance across her body and the surrounding walls, constantly shifting with her movements. The camera focuses on the tension in her muscles and the expressive gestures of her hands. A single, dramatic slow-motion shot captures her mid-air leap, with the light patterns swirling around her like a galaxy. Moody, artistic, high contrast.
Ghibli Forest Magic
Studio Ghibli-inspired animation with hand-painted watercolor textureStudio Ghibli-inspired anime style. A young girl with a straw hat lies peacefully in a sun-dappled magical forest, surrounded by friendly, glowing forest spirits (Kodama). A gentle breeze rustles the leaves of the giant, ancient trees. The air is filled with sparkling dust motes, illuminated by shafts of sunlight. The art style is soft, with a hand-painted watercolor texture. The scene feels serene, magical, and heartwarming.
完美適用於
技術規格
體驗原生音視頻生成
加入全球電影製作人、廣告商和創作者的行列,使用 Seedance 1.5 Pro 的突破性技術革新視頻內容創作。
Alibaba WAN 2.5 Image Edit
Alibaba WAN 2.5 Image Edit enables you to upload an existing visual and specify the desired adjustments. The model preserves layout and subject structure while implementing high-quality updates based on natural language.
Why creators love it
- Structure-preserving edits: Make lighting, color, or object changes without breaking composition.
- Text-guided styling: Reimagine materials, moods, or art styles with concise prompts.
- Prompt expansion on demand: Enable automatic prompt enrichment when you need extra detail.
- Flexible output sizes: Pick the resolution that best matches your downstream workflow.
Perfect for
- Marketing and design teams refining campaign visuals.
- E-commerce sellers upgrading product imagery.
- Content creators polishing thumbnails, covers, and posts.
- Artists experimenting with variations of their original work.
Billing rules
- Minimum charge: 1 image.
- Total cost = number of images × price per resolution.
How to use
- Provide the image you want to refine. (Image dimensions must be in (384, 5000))
- Describe the desired adjustments in the prompt.
- Choose the target resolution and submit.
- Review the enhanced output and download the version you like best.
Pro tips
- Start with clear instructions about colors, lighting, or objects to adjust.
- Pair positive and negative prompts to control what should or should not appear.
- Keep source images at or above your target resolution for optimal fidelity.
Note
If you did not upload the image locally, please ensure that the image URL is accessible! A successfully accessible image will display a preview in the interface.
Recommended Resolutions
| Aspect Ratio | Exact (W×H) | Exact Pixels | Rounded (W×H, ÷64) | Rounded Pixels |
|---|---|---|---|---|
| 1:1 | 1448 × 1448 | 2,096,704 | 1408 × 1408 | 1,982,464 |
| 3:2 | 1773 × 1182 | 2,095,686 | 1728 × 1152 | 1,990,656 |
| 4:3 | 1672 × 1254 | 2,096,688 | 1664 × 1216 | 2,023,424 |
| 16:9 | 1936 × 1089 | 2,108,304 | 1920 × 1088 | 2,088,960 |
| 21:9 | 2212 × 948 | 2,096,976 | 2176 × 960 | 2,088,960 |
| 1:1 | 1024 × 1024 | 1,048,576 | 1024 × 1024 | 1,048,576 |
| 3:2 | 1254 × 836 | 1,048,344 | 1216 × 832 | 1,011,712 |
| 4:3 | 1182 × 887 | 1,048,434 | 1152 × 896 | 1,032,192 |
| 16:9 | 1365 × 768 | 1,048,320 | 1344 × 768 | 1,032,192 |
| 21:9 | 1564 × 670 | 1,047,880 | 1536 × 640 | 983,040 |
| 1:1 | 323 × 323 | 104,329 | 320 × 320 | 102,400 |
| 3:2 | 397 × 264 | 104,808 | 384 × 256 | 98,304 |
| 4:3 | 374 × 280 | 104,720 | 448 × 320 | 143,360 |
| 16:9 | 432 × 243 | 104,976 | 448 × 256 | 114,688 |
| 21:9 | 495 × 212 | 104,940 | 576 × 256 | 147,456 |






