Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Alibaba Cloud / Wan 2.7

Wan 2.7 Text to Video · 2026-06-12

Wan 2.7 text-to-video with regional per-second prices for the 2026-06-12 snapshot.

availableClosed weightsVideo generation
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
35 of 43 specification fields collected
Output resolution720P, 1080PSee specification notes below
Clip duration2–15 secondsSee specification notes below
Input billingSee ratesAlibaba Model Studio · Singapore · International
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected
Maximum input tokens
Not collected
Maximum output
Not collected
Reasoning
Not collected
Streaming
Not collected
Knowledge cutoff
Not collected
Release date
2026-07-01Source for Release date (opens in a new tab)

Earliest reviewed regional platform release; snapshot suffix dates are not used as launch dates.

Parameters
Not collected
Architecture
Not collected
Languages
Chinese, EnglishSource for Languages (opens in a new tab)

Documented prompt languages; not a complete speech-output language roster.

API aliases
wan2.7-t2v-2026-06-12Source for API aliases (opens in a new tab)

Rolling names and dated snapshots remain separate records; exact checkpoint equivalence is not established.

Console model playground
SupportedSource for Console model playground (opens in a new tab)

Primary Singapore scope; regional availability can differ.

Maximum prompt characters
5,000Source for Maximum prompt characters (opens in a new tab)

Chinese characters and English letters count as characters; excess prompt text is truncated, not tokenized to this limit.

Visible watermark
Optional API watermark, off by defaultSource for Visible watermark (opens in a new tab)
Random seed range
0–2,147,483,647; omitted selects a random seedSource for Random seed range (opens in a new tab)

Identical seeds do not guarantee identical output.

Prompt rewriting
On by default; improves short prompts at the cost of added latencySource for Prompt rewriting (opens in a new tab)
Negative prompts
Up to 500 characters; longer text is truncatedSource for Negative prompts (opens in a new tab)
Request mode
Asynchronous task submission followed by polling; X-DashScope-Async: enableSource for Request mode (opens in a new tab)
Task ID lifetime
Task ID queryable for 24 hoursSource for Task ID lifetime (opens in a new tab)
Output link expiry
Generated-video URL expires after 24 hours; save the result promptlySource for Output link expiry (opens in a new tab)
Published availability regions
China (Beijing), SingaporeSource for Published availability regions (opens in a new tab)
Model revision
2026-06-12Source for Model revision (opens in a new tab)

Snapshot identifier date, not assumed to equal platform availability.

Generation modes
Text to video, Audio-guided video, Multi-shot narrativeSource for Generation modes (opens in a new tab)
Audio input limits
Public WAV/MP3 URL, 2–30 seconds, at most 15 MB; longer audio is trimmed to output duration, shorter audio leaves a silent tailSource for Audio input limits (opens in a new tab)
Aspect ratios
16:9 (default), 9:16, 1:1, 4:3, 3:4Source for Aspect ratios (opens in a new tab)
Source audio handling
Uses supplied audio or generates background music/sound effects when audio_url is omittedSource for Source audio handling (opens in a new tab)
Generation controls
Shot structure is controlled in the prompt; deprecated shot_type has no effectSource for Generation controls (opens in a new tab)
Output billing
Generated output duration only; input is not billed as video durationSource for Output billing (opens in a new tab)

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Alibaba Model Studio · Singapore · International

Provider source (opens in a new tab)
API model ID
wan2.7-t2v-2026-06-12
Access
POST https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis; poll /api/v1/tasks/{task_id}. Use an API key from the same region.
Regions
Singapore
Request quotas
300 requests per minuteSource for Request quotas (opens in a new tab)
Regional processing
International deployment scopeSource for Regional processing (opens in a new tab)
Free usage allowance
50 seconds; valid 90 days from latest activation, model release or application approvalSource for Free usage allowance (opens in a new tab)
Task submission rate
5 requests per secondSource for Task submission rate (opens in a new tab)
Quota scope
Primary account aggregate across RAM users, workspaces and API keys; each model has its own quotaSource for Quota scope (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Video generation720P output video1$0.10second
Video generation1080P output video1$0.15second

Rate conditions

  1. Generated output duration only; input is not billed as video duration. International deployment; native published USD list tariff, not a currency conversion. No text-token or per-question estimate.
Pricing source (opens in a new tab)

Data & enterprise terms

Task/result links expire after 24 hours. Full regional retention, training-use and enterprise contract terms are not verified by the API expiry statement.

Alibaba Model Studio · China (Beijing) · Chinese Mainland

Provider source (opens in a new tab)
API model ID
wan2.7-t2v-2026-06-12
Access
POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis; poll /api/v1/tasks/{task_id}. Use an API key from the same region.
Regions
China (Beijing)
Request quotas
300 requests per minuteSource for Request quotas (opens in a new tab)
Regional processing
Chinese Mainland deployment scopeSource for Regional processing (opens in a new tab)
Task submission rate
5 requests per secondSource for Task submission rate (opens in a new tab)
Quota scope
Primary account aggregate across RAM users, workspaces and API keys; each model has its own quotaSource for Quota scope (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Video generation720P output video1$0.082513second
Video generation1080P output video2$0.137521second

Rate conditions

  1. Generated output duration only; input is not billed as video duration. Chinese Mainland deployment; native published USD list tariff, not a currency conversion. Central pricing lists USD 0.086012/second, while this model card lists USD 0.082513/second. Confirm the account tariff before use. No text-token or per-question estimate.
  2. Generated output duration only; input is not billed as video duration. Chinese Mainland deployment; native published USD list tariff, not a currency conversion. Central pricing lists USD 0.143353/second, while this model card lists USD 0.137521/second. Confirm the account tariff before use. No text-token or per-question estimate.
Pricing source (opens in a new tab)

Data & enterprise terms

Task/result links expire after 24 hours. Full regional retention, training-use and enterprise contract terms are not verified by the API expiry statement.

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

Classification inference: this separately named native Wan 2.7/3.0 API release has no exact public checkpoint in the reviewed official Wan-AI inventory. Public Wan 2.1/2.2 models are excluded. API access and absence from that inventory do not independently prove undisclosed private checkpoint lineage; parameter counts and architecture are not transferred from public Wan models.

Classification evidence (opens in a new tab)

Lifecycle dates

Platform release · Singapore · International1 Jul 2026Source (opens in a new tab)
Platform release · China (Beijing)1 Jul 2026Source (opens in a new tab)

Sources reviewed