First and last frame control
Supply an opening frame, a closing frame, or both—the model plans the motion path while keeping the subject stable (observed Key features / FAQ).
Wan 2.7 AI Video Generator
Alibaba Tongyi WanxiangFirst/last framesMultimodal refs2–15s720p / 1080pVoice matchingFrom 160 credits
Wan 2.7 is Cinel’s keyframe-control Wan lane from Alibaba’s Tongyi Wanxiang team: write a prompt, lock a first and/or last frame, or guide with up to 3 images plus optional audio/video references, then render a 2–15 second clip at 720p or 1080p across five frame shapes. The advertised floor is from 160 credits (observed). Draft at 720p, then re-run keepers at 1080p. Choose Wan 2.7 when you need endpoint lock-ins and richer references—not Wan 2.5’s single-shot 480p floor, and not Wan 2.6’s first-frame-only multi-shot desk.
Keep Wan 2.6 for automatic multi-shot storytelling; keep Wan 2.5 for cheap single-shot tests.
From 160 credits · ~5s default · first/last frames + multimodal refs · not Seedance check-in silver
Examples
See Wan 2.7 in action
Still frames show keyframe-friendly moods—bioluminescent tidepool establishing shot, tactile paper-mech character beat, illustrated ink train leap. Live mp4 was not present on the public page at pack time (observed); if Mac later adds motion demos, keep them and use stills as backup.
Text-to-video mood: glowing tidepools hold; geodesic dome reflects moon ribbons over rocky shore
Reference/identity still→motion: brass-sage beetle keeps form while plates assemble on drafting table
Illustration motion: vermillion dusk train races bamboo corridor into ink mist
Why Wan 2.7
Why open Wan 2.7 on Cinel
Multimodal references
Up to 3 images plus member 15 s audio or video references, with optional 1–10 s voice matching (observed Specs / FAQ).
Flexible 2–15s at 720p/1080p
Ship five frame shapes with endpoint lock-ins—from 160 credits (observed).
Workflow
How to generate on Cinel
Step 1Upload keyframes (or start from text)
First frame, last frame, or both; or write a prompt (observed How to use).
Step 2Add references + pick length/aspect/resolution
Optional image/audio/video refs; start near 5s · 720p to verify cost (160 floor) (observed).
Step 3Generate, review timeline + voice, export
Confirm the live button price; raise to 1080p / longer only on keepers (observed).
Prompt Guide
Prompt tips for Wan 2.7
- Pair first and last frames that share lighting/style; describe the movement between them (observed).
- Write negative prompts as short comma-separated items (observed).
- Keep voice samples clean: single speaker, no music, within 1–10 s (observed).
- Draft at 720p; 1080p ≈ 1.67× and is best for finals (observed).
- Need automatic multi-shot storytelling → /wan-2-6.
- Need 480p single-shot floor → /wan-2-5.
Applications
What to ship with Wan 2.7
From precise ad transitions to recurring spokescharacters.
Precise ad transitions (open on product, end on pack shot)
Lock opening and closing keyframes so commercials land on exact brand guidelines.
Recurring spokescharacters with matched look + voice
Combine reference portraits with 1–10s audio samples for consistent character presentations.
Music-synced social clips from a short track
Upload member audio references to pace visual cuts and motion rhythms to music beats.
Fashion/beauty looks with multiple image refs
Feed up to 3 reference images to maintain complex outfits, styling and environment accents.
Model-level edits of short existing clips
Apply instruction-based edits to existing footage without re-generating from scratch.
Pricing & Credits
Credits and what “from 160” means
Wan 2.7 starts from 160 credits on Cinel for the default ~5 s · 720p combination. Duration (2–15 s), resolution (720p/1080p) and multimodal references change the total—the live figure is always shown on the Generate button (observed).
| Option / Tier | Cost / Multiplier | Details & Conditions |
|---|---|---|
| Default run (~5 s · 720p · MP4) | From 160 credits | Cinel baseline floor for Wan 2.7 keyframe generation (observed catalog). |
| Resolution ladder (720p · 1080p) | 1080p ≈ 1.67× per second | Draft at 720p; 1080p is ~1.67× the 720p rate per second (observed Specs). |
| Duration range (2–15 s) | Flexible 2–15s scaling | Adjust duration from 2 to 15 seconds to fit keyframe motion and scale proportionally (observed Specs). |
| Multimodal references & voice | Included or member-tiered | Up to 3 images; 15s audio/video references and 1–10s voice matching for members (observed Specs / FAQ). |
| Sibling model floors (observed) | 2.5: 75 · 2.6: 145 · 2.7: 160 | Wan 2.7 is Cinel’s dedicated keyframe & reference tier above 2.5 (budget single-shot) and 2.6 (multi-shot) (catalog observed). |
| Silver credits & check-in policy | Not Seedance 10 silver | Check-in +10 silver does not fund Wan 2.7 runs; no fake free claims. |
Transparent pricing with live calculation. Duration and resolution adjust your total in real time directly on the Generate button.
Deep Dive & Practical Guide
Wan 2.7 Capabilities & Guide
Explore generation speeds, resolution options, prompt fidelity, and workflow tips for Wan 2.7.
Wan 2.7 is the most controllable model in Alibaba's Tongyi Wanxiang lineup on Cinel. It can start from a prompt, animate between two keyframes you choose, or take guidance from a mix of reference images, a voice clip and a short video. Clips run 2 to 15 seconds at 720p or 1080p. If Wan 2.5 is the everyday clip and Wan 2.6 the storyteller, Wan 2.7 is the director's tool — for when you need the shot to start, look and sound a particular way. On Cinel it starts from 160 credits.
Wan 2.7 Feature Guide
What makes Wan 2.7 different
Three capabilities set it apart from its siblings. First, keyframe control: supply a first frame, a last frame or both, and the model works out the motion path between them while keeping the subject stable. Wan 2.6 only accepts a first frame.
Second, multimodal references. Beyond images, Wan 2.7 can take an audio clip and a reference video. On Cinel you can add up to 3 reference images to keep characters and visual details consistent, and the model can match a character's speech to the voice in a 1–10 second sample — useful for recurring hosts or spokescharacters.
Third, finer prompt control: negative prompts tell the model what to avoid (extra fingers, text overlays, a cluttered background), and optional prompt enhancement expands short ideas into fuller briefs. At model level Wan 2.7 also supports instruction-based editing of an existing 2–10 second clip, such as swapping the background for a rainy street without regenerating everything.
Wan 2.7 Feature Guide
Key features
- First and last frame control — plan exactly where the shot begins and ends.
- Image, audio and video references — up to 3 images plus a 15 s audio or video reference on Cinel (audio/video for members).
- Voice matching — character speech follows the vocal style of a short audio sample (model capability).
- Audio-driven timing — attach music or a voiceover and the motion is paced to it; if the track is shorter than the clip, the remainder is silent.
- Negative prompts and prompt enhancement for cleaner results.
- Any length from 2 to 15 seconds.
- 720p or 1080p (1080p ≈ 1.67× per second).
- Model-level video editing of existing short clips.
Wan 2.7 Feature Guide
Specs
| Item | Wan 2.7 |
|---|---|
| Developer | Alibaba — Tongyi Wanxiang team |
| Modes | Text-to-video, image-to-video with first/last frame, reference video; model-level video edit |
| Duration | 2–15 s (Cinel default 5 s) |
| Resolution | 720p / 1080p at 30 fps (1080p ≈ 1.67× per second) |
| Aspect ratios | 16:9, 9:16, 1:1, 4:3, 3:4 |
| References on Cinel | Up to 3 images · 15 s audio · 15 s video (audio/video member-only) |
| Voice sample | 1–10 s for voice matching |
| Prompt tools | Prompt up to 5,000 characters; negative prompt up to 500; optional prompt enhancement |
| Cinel cost | From 160 credits |
Wan 2.7 Feature Guide
How to use Wan 2.7 on Cinel
- Select Wan 2.7 in the video generator (video requires a membership or credit pack).
- Choose your control method: upload a first frame, a first and last frame, or a set of reference images.
- Members can add a short audio clip to set a character's voice or a video clip for motion and style guidance.
- Write the prompt, then add a negative prompt listing what to avoid — "no subtitles, no extra people, no warped hands".
- Set duration (2–15 s), aspect ratio and resolution; review at 720p, then render the keeper at 1080p.
Wan 2.7 Feature Guide
vs other Wan on Cinel
| Need | Choose |
|---|---|
| First/last-frame · multimodal refs · voice matching | Wan 2.7 (this page · from 160 credits) |
| Multi-shot · first-frame only · from 145 | Wan 2.6 |
| Single-shot · 480p tests · lowest Wan price | Wan 2.5 (from 75 credits) |
| Cross-model image→video picker | Image to Video |
Basic
Full access to all image & video models.
$9.99one-time
Permanent validity · Never expires600 Gold CreditsGood for everyday creation
Model capabilities
- Up to 600 Z-Image Turbo images
- Up to 120 GPT Image 2.5 (1K) images
- Up to 12 Seedance 1.0 Turbo AI videos — up to 12s · 1080p
◆ Unlimited & free generation
- FLUX.2 Klein 4BGold Member · Unlimited
- KolorsUnlimited · Free
- Up to 2 parallel outputs
- Up to 5 reference images
- No watermarks on any model
- Full commercial rights
- Priority generation queue
- Private creations by default
Plus
Full access to all image & video models.
$19.99one-time
Permanent validity · Never expires1,300 Gold CreditsBest value for creators
Model capabilities
- Up to 1,300 Z-Image Turbo images
- Up to 260 GPT Image 2.5 (1K) images
- Up to 27 Seedance 1.0 Turbo AI videos — up to 12s · 1080p
◆ Unlimited & free generation
- FLUX.2 Klein 4BGold Member · Unlimited
- KolorsUnlimited · Free
- Up to 4 parallel outputs
- Up to 10 reference images
- No watermarks on any model
- Full commercial rights
- High-speed VIP priority queue
- Private creations by default
Pro
Full access to all image & video models.
$49.99one-time
Permanent validity · Never expires3,500 Gold CreditsDeep volume discount
Model capabilities
- Up to 3,500 Z-Image Turbo images
- Up to 700 GPT Image 2.5 (1K) images
- Up to 72 Seedance 1.0 Turbo AI videos — up to 12s · 1080p
◆ Unlimited & free generation
- FLUX.2 Klein 4BGold Member · Unlimited
- KolorsUnlimited · Free
- Up to 4 parallel outputs
- Up to 10 reference images
- No watermarks on any model
- Full commercial rights
- High-speed VIP priority queue
- Private creations by default
Ultra
Full access to all image & video models.
$79.90one-time
Permanent validity · Never expires6,000 Gold CreditsLowest cost per credit
Model capabilities
- Up to 6,000 Z-Image Turbo images
- Up to 1,200 GPT Image 2.5 (1K) images
- Up to 125 Seedance 1.0 Turbo AI videos — up to 12s · 1080p
◆ Unlimited & free generation
- FLUX.2 Klein 4BGold Member · Unlimited
- KolorsUnlimited · Free
- Up to 4 parallel outputs
- Up to 10 reference images
- No watermarks on any model
- Full commercial rights
- High-speed VIP priority queue
- Private creations by default
FAQ
Before you create
What can Wan 2.7 do that Wan 2.6 cannot?
Wan 2.7 adds last-frame control, audio and video references, voice matching from a short sample, negative prompts and model-level video editing. Wan 2.6 supports a first frame only (observed).
How do first and last frames work in Wan 2.7?
Upload one image for the opening frame, one for the closing frame, or both. The model infers the movement between them and keeps the subject consistent (observed).
Can Wan 2.7 copy a voice?
In reference mode the model can match a character's speech to the vocal characteristics of a 1–10 second audio sample. Only use voices you have permission to use (observed).
How many references can I add on Cinel?
Up to 3 reference images, plus one audio or video reference of up to 15 seconds for members (observed).
Which aspect ratios does Wan 2.7 support?
Five: 16:9, 9:16, 1:1, 4:3 and 3:4 (observed).
How many credits does Wan 2.7 cost on Cinel?
It starts from 160 credits. Duration, resolution and references raise the estimate displayed before you generate (observed).
Related Models & Tools
Related AI Creation Experiences
Explore automatic multi-shot storytelling, everyday single-shot Wan, or the cross-model image-to-video hub.
Ready to lock first and last frames with Wan 2.7?
Open the generator, upload your endpoints or references, check the live credit estimate, and render.