Create Wan 2.7 videos from text, images, video references, or audio direction. Use multi-shot prompts, first/last-frame transitions, continuation scenes, and synchronized sound for short cinematic clips at 720p or 1080p.
Wan 2.7 is Alibaba's video generation model family for text-to-video, image-to-video, and reference-to-video creation. It is built for short clips that need stronger prompt following, reference-guided subjects, multi-shot structure, camera direction, and synchronized audio cues.
Use Wan 2.7 when a single prompt needs to describe the scene, reference media, shot order, first and last frames, continuation timing, motion style, lighting, and sound. It fits product ads, ASMR material clips, social videos, storyboard tests, character motion, and reference-led commercial drafts.
AI Video Model
| Performance Metric | Toolaze Specification |
|---|---|
| Creation modes | Text-to-video, image-to-video, and reference-to-video |
| Reference media | Up to five total image or video references in reference-to-video; audio references where supported |
| Frame control | First-frame, first-and-last-frame, and video continuation prompts |
| Shot control | Single-shot clips, multi-shot sequences, camera movement, shot scale, and scene timing |
| Audio control | Audio-video synchronization, sound effects, music, ambience, and voice-style direction |
| Output window | 2 to 15 second official duration range with 720p or 1080p output |
| Toolaze Credits | Starts at 64 Credits for 720p 2-second output; selected duration and resolution update the Credits requirement before generation |
Open the AI video generator and select Wan 2.7 when you want the newest Wan family video model for prompt-driven or reference-guided clips.
Start from text, an image, first and last frames, a video continuation idea, or a set of image, video, and audio references when the mode supports them.
Describe the subject, reference roles, shot order, camera movement, timing, lighting, and sound in one clear prompt.
Pick the available aspect ratio, duration, and resolution, generate the clip, then adjust the prompt if motion, identity, or sound timing needs another pass.
Copy these examples when you need multi-shot motion, reference control, first/last-frame transitions, or synchronized sound.
Multi-Shot Product Launch
Create a Wan 2.7 video for a matte black wireless speaker. Shot 1: slow push-in on the speaker on a concrete table. Shot 2: close-up of the fabric grille vibrating with bass. Shot 3: water droplets slide across the surface. Shot 4: the speaker sits in a warm living room at sunset. Keep the product shape consistent, add soft camera motion, clean reflections, deep bass pulses, and subtle room ambience.
First and Last Frame Transition
Animate from the uploaded first frame of a closed sketchbook on a desk to the uploaded last frame of the same sketchbook open with a finished city map. The camera stays top-down, pages turn naturally, pencil lines draw themselves across the paper, desk objects remain fixed, and the final map lands exactly on the last reference frame with soft paper rustle.
Reference-to-Video Character Scene
Use three references: a character design sheet, a rainy alley background, and a short neon sign video. Create a Wan 2.7 clip where the character walks toward the camera, stops under the sign, and looks up as rain reflects on the coat. Preserve the face, outfit, colors, and alley layout. Add slow side-tracking camera movement, footsteps, rain on metal, and a low city hum.
Continuation With Native Sound
Continue the uploaded cafe counter video. Keep the same framing, lighting, cup position, and barista hands. The milk stream starts thin, becomes a smooth spiral, and finishes as a leaf pattern on the latte. Add synchronized pouring sound, ceramic cup contact, quiet machine hum, and no sudden camera jump.
Stronger Wan 2.7 prompts tell the model what changes, what stays fixed, and how the camera and sound should move over time.
Weak prompt
A cool sneaker commercial.
Better prompt
Create a sneaker commercial. Shot 1: macro close-up of the sole touching wet pavement. Shot 2: side tracking shot as the runner accelerates. Shot 3: slow push-in on the shoe logo under neon reflections. Keep the same sneaker design, add splash sound, city ambience, and crisp lighting.
Weak prompt
Use these images and make a video.
Better prompt
Use image 1 for the character face and outfit, image 2 for the room layout, and video 1 for the neon light motion. Create a clip where the character turns toward the window while the neon reflection moves across the wall. Preserve identity, clothing colors, and room geometry.
Weak prompt
Make ASMR with good audio.
Better prompt
Create a macro ASMR clip of a glass fruit being sliced. The blade enters early, the fruit cracks midway, and tiny pieces slide apart before the ending. Add crisp glass tapping, soft cutting pressure, quiet table contact, and no background music.
Plan short commercial sequences with product close-ups, reveal shots, camera moves, and synced material sounds.
Animate character sheets, outfit references, and background references while keeping identity and style stable.
Move from one planned frame to another for transformations, reveals, before-after visuals, and story beats.
Extend a short source clip while preserving framing, lighting, subject position, and motion direction.
Create tactile cuts, pours, taps, fabric motion, machine hums, rain, room tone, and other synchronized audio moments.
Draft vertical hooks, cinematic beats, campaign ideas, creator intros, and mood loops before producing final assets.
Compare the features that change the result: frame control, references, audio, duration, resolution, and credits.
| Feature | Wan 2.7 | Wan 2.6 | Wan 2.5 |
|---|---|---|---|
| Text-to-video | |||
| Image-to-video reference | One first frame plus an optional last frame. | One uploaded image. | One uploaded still image. |
| Start/end frame control | |||
| Continuation control | First-frame, first-and-last-frame, and video continuation-style planning. | ||
| Audio handling | Audio-video synchronization cues in prompts. | Native audio output with synchronized sound cues. | Native sound cues for simpler clips. |
| Duration | 2 to 15 seconds. | 5, 10, or 15 seconds. | 5 or 10 seconds. |
| Resolution | 720p or 1080p. | 720p or 1080p. | 720p or 1080p. |
| Starting Credits | Starts at 64 Credits for 2-second 720p output. | Starts at 140 Credits for 5-second 720p output. | Starts at 120 Credits for 5-second 720p output. |
| Choose it when | You need the start, ending frame, or transition path to be explicit. | You need a stable longer Wan clip from text or one image. | You need quick drafts and simple one-image motion tests. |
Use this comparison to choose the model that matches your input material, motion needs, sound direction, and output target.
| Model choice | Good fit | Check before generating |
|---|---|---|
| Wan 2.7 | Reference-led Wan clips with text, images, video references, audio direction, multi-shot prompts, first/last-frame control, and 720p/1080p output. | Choose it when reference control, frame transitions, continuation, and synchronized sound matter more than maximum resolution. |
| Kling 3.0 | Dynamic cinematic movement, stronger high-resolution options, multi-image input, and expressive action scenes. | Choose it when fast camera movement, 4K output, or dramatic motion is the main goal. |
| Seedance 2.0 | Director-style multi-shot video generation with strong camera planning, character movement, and multiple references. | Choose it when longer scene planning, multiple reference types, or commercial shot structure is more important than Wan-specific controls. |
| Grok Imagine Video 1.5 | Fast image-to-video drafts, creator clips, product motion, and lightweight 480p/720p tests. | Choose it when speed and low-friction iteration are more important than detailed reference control. |
Feedback from creators using Wan 2.7 when the first frame, last frame, transition path, and final composition all matter.
Maya Chen
Product video editor
“I use Wan 2.7 when a product reveal needs to land on a specific final composition instead of drifting into a random ending. The last-frame anchor makes client review easier because everyone can see the intended destination.”
Eli Morgan
Storyboard artist
“The first and last frame setup helps me show clients the exact change I want before building a longer piece. I can sketch the start, define the landing frame, and use the prompt to describe the motion path between them.”
Nora Patel
Retail creative lead
“For packaging and display transformations, I write the transition path and keep the final frame as the anchor. Wan 2.7 is useful when the shelf layout, product facing, and end card need to stay intentional.”
Jon Bell
Campaign motion director
“Wan 2.7 is the version I choose when the end state matters as much as the motion. It lets me test a campaign beat that starts from a clean hero frame and resolves into a planned product lockup.”
Clara Evans
Social launch editor
“When a launch clip has to open on one visual and end on a precise reveal, Wan 2.7 saves me from guessing. I can keep the hook short while still controlling where the viewer lands.”
Owen Miller
Performance creative strategist
“For paid creative tests, I use Wan 2.7 when we already know the start and final product message. The first-and-last-frame control gives us cleaner variants for comparing transition speed, framing, and final recall.”
Wan 2.7 is Alibaba's video generation model family for creating short videos from text, images, video references, and audio direction. On Toolaze, it is presented as a browser-based Wan 2.7 generator with prompt, output, and reference controls.
Wan 2.7 supports text-to-video, image-to-video, and reference-to-video. Reference-to-video can use image, video, and audio references to guide the subject, scene, motion, and sound.
Yes. Wan 2.7 can use image references, including first-frame and first-and-last-frame style control. For reference-to-video, Wan 2.7 supports up to five total image or video references in supported modes.
Yes. Wan 2.7 reference-to-video can use video references, and Wan 2.7 image-to-video includes video continuation so a source clip can continue with matching framing, lighting, and motion direction.
Yes. Wan 2.7 supports audio-video synchronization for effects, ambience, music, and voice-style direction. For best results, write sound as timed scene action instead of adding a vague audio request.
Yes. Wan 2.7 supports multi-shot prompting. Write the sequence as Shot 1, Shot 2, and Shot 3, then describe what changes and what should remain consistent across the clip.
Wan 2.7 supports a 2 to 15 second duration range with 720p or 1080p output. Toolaze shows the available duration and resolution choices in the generator before you create the video.
Include the subject, reference roles, motion, camera movement, shot order, timing, lighting, and sound. Short, structured prompts usually work better than broad cinematic descriptions.
Wan 2.7 starts at 64 Credits for a 720p, 2-second generation on Toolaze. Longer duration or 1080p output increases the required Credits before generation.
Wan 2.7 video generation depends on your available Credits, account state, selected duration, and selected resolution. The generator shows the current Credits requirement before generation.
Choose Wan 2.7 when you need the newest Wan controls, richer references, first/last-frame transitions, video continuation, multi-shot prompts, and synchronized sound. Choose Wan 2.6 or Wan 2.5 for simpler Wan clips with fewer setup choices.
Choose Kling when high-resolution action or dramatic motion is the priority, Seedance when director-style scene planning is more important, and Grok Imagine Video when you want a faster draft from an image.
Use Seedance 2.0 for director-style multi-shot video generation and reference-heavy commercial scenes.
Try Kling 3.0 for dynamic motion, cinematic camera movement, and high-resolution video output.
Use Grok Imagine Video 1.5 for quick image-to-video drafts, product motion, and creator clips.