Wan is Alibaba's multimodal generation suite covering both image and video workflows. Beyond classic 2.6 video modes (T2V, I2V, V2V), Wan now supports 2.7 Image / Image Pro and four advanced 2.7 video modes: v2.7 Text to Video, v2.7 Image to Video, v2.7 Video Edit, and v2.7 R2V for richer control and production-ready outputs.
Generate or edit images with model wan-2-7-image. Supports optional input_urls (up to 9), optional aspect ratio when no image input, sequential batch mode, n control, 1K/2K resolution, optional thinking mode (eligible conditions), color palette, bbox list, watermark, and seed.
Use this feature →Pro image generation/editing with model wan-2-7-image-pro. Includes all Wan 2.7 Image controls plus 4K resolution support when no input image and sequential mode is off.
Use this feature →Generate video directly from text prompts. Supports Chinese and English, 1–5000 characters. Choose duration (5, 10, or 15 seconds), resolution (720p or 1080p), and shot composition (single continuous shot or multiple shots with transitions).
Use this feature →Animate static images with text guidance. Provide image URLs (min 256×256px; JPEG, PNG, WebP; max 10MB), add a prompt describing the motion or scene, and set duration, resolution, and multi-shot style. Perfect for bringing illustrations, product shots, or concept art to life.
Use this feature →Transform existing video with new prompts. Upload a video URL (MP4, MOV, MKV; max 10MB), describe the desired changes or style transfer in text, and control duration (5 or 10 seconds), resolution, and shot composition. Ideal for restyling, content adaptation, or creative remixing.
Use this feature →Model fixed to wan-2-7-text-to-video. Required prompt (3–5000 chars), optional negative prompt/audio URL, 720p/1080p, ratio 16:9/9:16/1:1/4:3/3:4, duration 2–15s, prompt_extend, watermark, and seed.
Use this feature →Model fixed to wan-2-7-image-to-video. Required prompt (3–5000); at least one of first_frame_url, last_frame_url, first_clip_url, driving_audio_url; optional negative prompt; 720p/1080p, duration 2–15s, prompt_extend, watermark, seed.
Use this feature →Model fixed to wan-2-7-videoedit. Required videoUrl and prompt (3–5000); optional negativePrompt/referenceImage; 720p/1080p, optional aspectRatio, duration 0 or 2–10s, audioSetting (auto/origin), promptExtend, watermark, seed.
Use this feature →Model fixed to wan-2-7-r2v. Required prompt (3–5000); referenceImage[] and/or referenceVideo[] (total ≤ 5); optional firstFrame (with it aspectRatio omitted), referenceVoice; 720p/1080p, aspectRatio, duration 2–10s, promptExtend, watermark, seed.
Use this feature →Motion meets imagination: Wan bridges the gap between static concepts and dynamic storytelling. Whether you're starting from text, an image, or existing footage, Wan gives you the tools to craft video content with the same ease as generating an image.
Flexible, fast, and accessible: With support for both Chinese and English, multiple duration options, and resolution choices from web-ready to HD, Wan adapts to your workflow—from quick social clips to polished creative projects.
Wan 2.7 Image
Model: Fixed to wan-2-7-image.
Prompt: Chinese/English supported, up to 5000 characters.
Input: Optional input_urls (up to 9 images). If input images are provided, aspect ratio is hidden and not sent.
Aspect ratio (no input image): 1:1, 16:9, 4:3, 21:9, 3:4, 9:16, 8:1, 1:8.
Sequential mode: enable_sequential false by default. When false, n range is 1–4 (default 4). When true, n range is 1–12 (default 12).
Resolution: 1K or 2K.
Thinking mode: Available only when enable_sequential=false and no input_urls.
Advanced controls: optional color_palette (3–10 colors, only when non-sequential), optional bbox_list for interactive edit regions, watermark toggle, seed 0–2147483647.
Use cases: poster generation, product visual variants, style-consistent batch assets, guided local edits.
Wan 2.7 Image Pro
Model: Fixed to wan-2-7-image-pro.
Same controls as Wan 2.7 Image, with extended resolution support: 1K / 2K / 4K.
4K constraint: valid only when no input image and sequential mode is disabled.
Use cases: high-resolution campaign key visuals, premium product renders, print-friendly hero assets.
Text to Video
Prompt: Natural language in Chinese or English, 1–5000 characters.
Duration: 5 seconds (quick loops), 10 seconds (standard clips), or 15 seconds (extended scenes).
Resolution: 720p (fast, web-optimized) or 1080p (Full HD, higher detail).
Shot composition: Single continuous shot for simplicity, or multi-shot with transitions for narrative flow.
Use cases: Concept visualization, ad creatives, social media content, storyboarding.
Image to Video
Input: Image URL(s) required; minimum 256×256px; formats JPEG, PNG, WebP; max 10MB per image.
Prompt: Describe the animation, motion, or scene evolution (1–5000 characters).
Duration: 5, 10, or 15 seconds.
Resolution: 720p or 1080p.
Shot composition: Same multi-shot controls as text-to-video.
Use cases: Animating illustrations, product demos, character movement, bringing concept art to life.
Video to Video
Input: Video URL required; formats MP4, MOV, MKV; max 10MB.
Prompt: Describe the transformation—style transfer, content change, mood shift, etc. (1–5000 characters).
Duration: 5 or 10 seconds (output length may differ from input).
Resolution: 720p or 1080p.
Shot composition: Single or multi-shot, with ability to reinterpret original footage.
Use cases: Restyling existing content, adapting videos for different platforms, creative remixing, consistent brand video generation.
Image campaign production: Build social posters, ad variants, product galleries, and style-matched visual sets with Wan 2.7 Image / Image Pro.
High-res hero assets: Use Wan 2.7 Image Pro (4K in eligible conditions) for launch banners, e-commerce covers, and high-impact brand creatives.
Social media content: Generate short videos for TikTok, Instagram Reels, YouTube Shorts—in the right duration and resolution.
Advertising and marketing: Create product demos, brand stories, and promotional clips from text briefs or existing assets.
Creative projects: Animate illustrations, bring storyboards to life, experiment with video styles and transitions.
Educational content: Produce explainer videos, visual aids, and short tutorials without complex editing software.
Concept validation: Quickly visualize video ideas before committing to full production.
Content adaptation: Transform existing videos for different audiences, platforms, or brand guidelines.
Text-to-video generation: 30–90 seconds for 5s clips, 60–180 seconds for 15s clips (dependent on resolution and shot complexity).
Image-to-video: 40–120 seconds; image complexity and desired motion affect speed.
Video-to-video: 60–180 seconds; depends on input length, transformation complexity, and resolution.
Concurrency: Supports 50+ parallel requests with intelligent queue management.
Prompt length: Up to 5000 characters across all modes.
Input formats: JPEG, PNG, WebP for images; MP4, MOV, MKV for videos (max 10MB each).
Output: MP4 format, delivered via secure URL or direct download.
Image workflow (Wan 2.7 Image / Pro): Write prompt → (optional) upload input image(s) → set n / resolution / sequential / optional advanced controls → generate → pick best outputs.
Quick create (text-to-video): Write prompt → choose duration/resolution/shot style → generate → preview → iterate with refined prompt or parameters.
Animate assets (image-to-video): Upload image → describe desired motion → set parameters → generate multiple variations → select best animation.
Transform footage (video-to-video): Upload video → describe transformation → choose duration/resolution → generate → review → refine prompt if needed.
Batch production: Plan content calendar → generate multiple videos with consistent parameters → unify style across campaign assets.
Image prompt crafting (2.7 Image / Pro): Specify subject, style, composition, and constraints; for edit tasks, clearly state what to change and what to preserve.
Prompt crafting: Be specific about motion, scene changes, and style. Example: "A product rotating slowly on a white background, studio lighting, smooth 360-degree view" vs. just "product video."
Duration strategy: Use 5s for loops and quick social clips; 10–15s for storytelling or demonstrations.
Resolution choice: 720p for drafts, quick reviews, and web-first content; 1080p for final assets, presentations, and HD platforms. For 2.7 image workflows, use 1K/2K by default; use 4K (Pro only) when no input image and sequential is off.
Shot composition: Single shot works best for focused subjects; multi-shot adds narrative depth—use when you need scene changes or progression.
Image quality: For image-to-video, higher resolution inputs (ideally 1024×1024 or larger) yield smoother animations.
Video-to-video prompts: Be explicit about what to change—style, mood, objects, background—and what to preserve.
Cost efficiency: Match duration and resolution to platform requirements; batch similar requests to reuse parameters.
Unified image + video stack: Generate high-quality visuals with Wan 2.7 Image / Pro, then extend the same idea into motion with Wan video models.
Multilingual support: Generate from prompts in Chinese or English—seamless for global teams.
Flexible input modes: Text, image, or video—start from wherever your creative process begins.
Duration control: 5, 10, or 15 seconds—fit any platform's requirements.
HD quality: Up to 1080p for video outputs and up to 4K (Wan 2.7 Image Pro in eligible scenarios) for image outputs.
Shot composition: Single or multi-shot—choose the right narrative structure.
Ease of use: Intuitive parameters, no video editing expertise required.
Best for: Social media managers, marketers, designers, content creators, educators, advertisers, and anyone needing fast, high-quality image + video generation.
Wan on FuseAITools makes both image and video creation as simple as writing a sentence. Start with Wan 2.7 Image / Image Pro for key visuals, then move into T2V/I2V/V2V for motion outputs. With controls for ratio, batch size, sequential mode, quality, and advanced edits, Wan is a practical all-in-one visual generation stack for modern teams.