Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Akool / Akool visual AI

Talking Photo

Speech-driven photo animation with a gesture prompt and selectable video resolution.

availableClosed-source managed componentVideo generation
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
3 of 17 specification fields collected
Output resolution720p, 1080pSee specification notes below
Clip durationNot collectedAwaiting source review
Input billingSee ratesAkool
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

No model-specific public value established in reviewed primary sources.

Maximum output
Not collected

No model-specific public value established in reviewed primary sources.

Context caching
Not collected

No model-specific public value established in reviewed primary sources.

Reasoning
Not collected

No model-specific public value established in reviewed primary sources.

Tool calling
Not collected

No model-specific public value established in reviewed primary sources.

Structured output
Not collected

No model-specific public value established in reviewed primary sources.

Streaming
Not collected

No model-specific public value established in reviewed primary sources.

Batch processing
Not collected

No model-specific public value established in reviewed primary sources.

Fine-tuning
Not collected

No model-specific public value established in reviewed primary sources.

Parameters
Not collected

No model-specific public value established in reviewed primary sources.

Architecture
Not collected

No model-specific public value established in reviewed primary sources.

Knowledge cutoff
Not collected

No model-specific public value established in reviewed primary sources.

Languages
Not collected

No model-specific public value established in reviewed primary sources.

Release date
Not collected

No model-specific public value established in reviewed primary sources.

Input modalities
Image, Audio, TextSource for Input modalities (opens in a new tab)
Output resolutions
720p, 1080pSource for Output resolutions (opens in a new tab)

Default 720p. Text prompt controls gestures rather than generating speech.

Access & pricing

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

API model ID
Not collected

The route, job ID, voice ID or avatar ID does not identify a native serving checkpoint.

Access
POST https://openapi.akool.com/api/open/v3/content/video/createbytalkingphoto. Authenticate with x-api-key or Bearer authorization. Account, plan and feature availability apply.
Regions
Published personal-data storage: AWS in the EEA or United States unless otherwise specified. This is not a guarantee of a selectable inference region.
Training data policy
Published terms prohibit use of customer content or usage data to train machine-learning models.Source for Training data policy (opens in a new tab)
Data retention
Customer Content is stored for 365 days and users can clear it; other data categories and contractual/legal exceptions have separate retention rules.Source for Data retention (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Published credit rateTalking Photo video110 credits5 seconds

Rate conditions

  1. Published Pro Max and Business rate; Enterprise custom.
Pricing source (opens in a new tab)

Data & enterprise terms

Terms state customer content and usage data will not be used by Akool or third parties to train machine-learning models. Privacy policy gives Customer Content a 365-day retention period with user deletion, and AWS storage in the EEA or US unless otherwise specified. Generated-resource expiry is separate from storage retention. Contract-specific and legal exceptions apply.

Read the service policy (opens in a new tab)

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.

Akool

Not verified

Saved media, avatar templates, voice assets and temporary result URLs do not establish prompt caching or discounted cached-input pricing.

How it works
Not collected

Saved media, avatar templates, voice assets and temporary result URLs do not establish prompt caching or discounted cached-input pricing.

Minimum reusable input
Not collected

Saved media, avatar templates, voice assets and temporary result URLs do not establish prompt caching or discounted cached-input pricing.

Lifetime & refresh
Not collected

Saved media, avatar templates, voice assets and temporary result URLs do not establish prompt caching or discounted cached-input pricing.

How to enable it
Not collected

Saved media, avatar templates, voice assets and temporary result URLs do not establish prompt caching or discounted cached-input pricing.

Usage fields
Not collected

Saved media, avatar templates, voice assets and temporary result URLs do not establish prompt caching or discounted cached-input pricing.

Billing rules
Not collected

Saved media, avatar templates, voice assets and temporary result URLs do not establish prompt caching or discounted cached-input pricing.

Limits & data boundaries
Not collected

Saved media, avatar templates, voice assets and temporary result URLs do not establish prompt caching or discounted cached-input pricing.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 14 Oct 2026

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

Documented named hosted service, product or voice-engine selector. The underlying native checkpoint is undisclosed or explicitly supplied by a third party, so this record does not assert independent proprietary Akool weights.

Classification evidence (opens in a new tab)

Sources reviewed