curl-X POST 'https://pollo.ai/api/platform/v1/generation/alibaba/wan-v2-6/video'\-H'Content-Type: application/json'\-H'x-api-key: YOUR_API_KEY'\-d'{ "input": { "prompt": "A cinematic shot of a golden retriever running through a field of sunflowers at sunset, warm rim light, shallow depth of field", "negativePrompt": "blurry, low quality, distorted, watermark, text", "duration": 5, "resolution": "720p", "aspectRatio": "1:1", "generateAudio": true, "multiShot": false }}'
Optionally include webhookUrl to receive a callback when the task succeeds or fails.
Optional request field
{"input":{"prompt":"A cinematic shot of a golden retriever running through a field of sunflowers at sunset, warm rim light, shallow depth of field","negativePrompt":"blurry, low quality, distorted, watermark, text","duration":5,"resolution":"720p","aspectRatio":"1:1","generateAudio":true,"multiShot":false},"webhookUrl":"https://example.com/webhooks/pollo"}
Schema
Input fields for the selected endpoint and mode.
Field
Type
Required
Description
audio
string
No
Reference audio URL (HTTP(S)).
prompt
string
Yes
Text prompt describing the content to generate.
negativePrompt
string
No
Text describing elements to avoid in the generated output.
duration
integer
No
The duration of the generated video, in seconds.
resolution
string
No
Output resolution of the generated media (e.g. 720p, 1080p, 2K, 4K). Values are case-insensitive.
aspectRatio
string
No
Output aspect ratio, as width:height (e.g. 16:9).
seed
integer
No
Random seed for reproducible generation.
generateAudio
boolean
No
Generate natural-sounding, fitting audio for the output video.
multiShot
boolean
No
Generate a multi-shot video (multiple scene cuts with automatic transitions) instead of a single continuous shot.
Wan 2.6 supports multi-shot video generation for short-form storytelling. It can connect several shots in one clip while helping maintain character appearance, visual style, and motion consistency.
On Pollo API, developers reach Wan 2.6 for text-to-video and image-to-video generation with matching audio, clip lengths up to 15 seconds, and 720P or 1080P output per request.
Key Features of Wan 2.6 API
Multi-Shot Sequences
Rather than one continuous take, Wan 2.6 can chain several shots into a single generation, letting a short story move between angles and moments without breaking continuity.
Extended Clip Lengths
Clips run anywhere from two to fifteen seconds, leaving room for a setup, a reaction, and a payoff that shorter models tend to cut off before the idea lands.
Synced Audio Output
The model can generate fitting audio alongside the visuals or follow an audio track you supply, so sound and motion line up instead of arriving as two separate steps.
Character and Style Consistency
Guided by visual references, Wan 2.6 keeps a subject's look and the overall art direction stable across shots, which matters when a face or product recurs through a sequence.
Flexible Framing
Every aspect ratio is available at both resolutions, so square, portrait, and widescreen outputs render at full detail without trading shape against quality.
Reliable Scene Logic
Described actions, shot order, and staging land closely, so the sequence follows the intended logic instead of drifting into unrelated motion or composition.
Use Cases of Wan 2.6 API
Mini-Narrative Ads: Build short story-driven spots that move across two or three shots, carrying a product or idea from tease to resolution in one clip.
Explainer and Tutorial Clips: Show a process one stage per shot, with fitting narration audio, for how-to tools and onboarding content.
Music and Sound-Led Scenes: Supply an audio track and let the visuals follow its rhythm, useful for lyric clips, promos, and mood pieces.
Product Sequences: Take an item from setup to use across several angles while its shape, color, and finish stay consistent throughout.
Character-Driven Shorts: Keep a recurring character looking the same across shots for episodic, serialized, or mascot-based content.
Image-to-Motion Stories: Animate a single still into a longer moving sequence with sound, extending one asset into a full moment.
Social Verticals: Render portrait clips up to fifteen seconds for feed-native storytelling that finishes an idea within a single scroll.
How to Use Wan 2.6 API
Get an API Key: Create a Pollo API account and generate your API key from the developer dashboard.
Choose the Model: Send requests to the Wan 2.6 endpoint at /generation/wanx/wan-v2-6.
Add Your Inputs: Provide a prompt for text-to-video or an image URL for image-to-video, then set length (2 to 15), resolution (720P or 1080P), and optionally audioUrl, generateAudio, or multiClip.
Generate and Retrieve: Submit the task, poll the returned task status, and download the finished video once processing succeeds.
Prompting Best Practices for Wan 2.6 API
Prompt Wan 2.6 like a shot list. Name each moment in order, describe the subject and its motion, and state the sound, so multi-shot pacing and audio have a clear target.
A Simple Prompt Formula
Subject and setting + shot-by-shot actions in order + motion and pacing + audio or sound cue + style and framing
What You Should Notice
Sequence Your Shots: Spell out what happens first, next, and last so the model knows where each shot begins and ends.
Anchor The Subject: Describe a consistent look for any character or product that reappears across shots.
Call The Sound: Say what should be heard, whether ambient noise, a voice, or music matched to the action.
Set The Pace: Longer clips need clear beats, so note where the moment slows down or speeds up.
Match Resolution To Use: Choose 1080P for detail-heavy sequences and 720P for faster drafts and iteration.
Fix The Frame: Pick an aspect ratio that fits the destination, since all shapes render at both resolutions.
Example Prompts
Multi-Shot Cooking Segment
"Three shots: hands crack eggs into a bowl, a whisk beats the mixture into froth, then batter pours onto a hot pan and bubbles. Warm kitchen light, close overhead framing, gentle sizzling audio, clean realistic style."
Museum Walkthrough
"A visitor walks slowly past tall dinosaur skeletons, camera tracking beside them, then cuts to a low angle looking up at the ribcage. Soft echoing footsteps and quiet ambient hall audio, natural lighting, documentary realism."
Audio-Led Street Musician
"A cellist plays on a stone bridge at dusk, bow moving with the melody, then the shot widens to reveal a small gathering crowd. Warm string audio synced to the motion, golden hour glow, handheld framing."
Wildlife Sequence
"A fox steps out of tall grass, ears turning, then bounds forward as the camera pans to follow across a misty field. Rustling grass and distant birdsong audio, cool morning light, grounded natural style."
Wan 2.6 vs Sora 2 vs Kling 2.6
Capability
Wan 2.6
Sora 2
Kling 2.6
Multi-shot sequences
✅ Native shot chaining
✅ Strong scene stitching
➖ Mostly single directed shots
Clip length range
2 to 15 seconds
Short fixed durations
Short directed clips
Generated or audio-driven sound
✅ Both supported
✅ Generated audio
❌ Limited native audio
Aspect ratio flexibility
✅ All ratios at both resolutions
✅
✅
Resolution options
720P and 1080P
Up to 1080P
Up to 1080P
Character consistency
✅ Reference-guided
✅
✅
Recommended For
Longer multi-shot stories with audio
Coherent generated scenes
Character motion and detail
Why Choose Wan 2.6 API?
Wan 2.6 fits products that need more than a single clip, carrying a character or idea across multiple shots with matched audio and clip lengths that stretch to fifteen seconds.
Through Pollo API, you reach Wan 2.6 with one API key alongside 300+ leading video and image models, with clean docs, task status polling, and faster generation.
Start with the Wan 2.6 endpoint on Pollo API to ship longer, sound-synced storytelling features, then compare it against Sora 2, Kling, and Veo without touching your integration.
Wan 2.6 API FAQs
What is Wan 2.6?
Wan 2.6 is a wanx video generation model built for multi-shot storytelling, supporting text-to-video and image-to-video with synced audio, consistent characters, and clips from two to fifteen seconds.
Does Wan 2.6 API support text and image inputs?
Yes. You can generate from a text prompt or animate an image URL, making it flexible for both idea-first and asset-first workflows.
Can Wan 2.6 create multi-shot videos?
Yes. It can chain several shots into one clip while keeping character and style consistent, so a short story moves between moments without visible breaks.
How does audio work in Wan 2.6?
The model can generate fitting audio for a clip or follow an audio track you provide, aligning sound with the on-screen motion in the same request.
What clip lengths and resolutions are available?
Wan 2.6 supports clip lengths from two to fifteen seconds at 720P or 1080P, with every aspect ratio available at both resolutions.
Why run Wan 2.6 through Pollo API?
Pollo API gives one integration for Wan 2.6 and 300+ other models, with clear documentation, task tracking, and lower-cost generation than comparable providers.
