Pruna AI / P-Image

P-Image-Try-On

Managed garment try-on workflow combining a person image and garment references, with optional pose guidance and faster processing.

Image generation

availableClosed-source managed componentImage editing

Official documentation (opens in a new tab)Reviewed 15 Oct 2026
5 of 21 specification fields collected

Output formatsWebP, JPEG, PNGSee specification notes below

Reference imagesUp to 11 garment references; up to 6 recommended, plus the person image and optional pose reference.See specification notes below

Input billingSee ratesPruna API

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
Maximum prompt characters: Not collected
Reasoning: Not collected
Tool calling: Not collected
Structured output: Not collected
Streaming: Not collected
Batch processing: Not collected
Fine-tuning: Not collected
Context caching: Not collected
Knowledge cutoff: Not collected
Release date: Not collected
Parameters: Not collected
Architecture: Not collected
Languages: Not collected
Input modalities: Image, TextSource for Input modalities (opens in a new tab)
Output modalities: ImageSource for Output modalities (opens in a new tab)
Reference images: Up to 11 garment references; up to 6 recommended, plus the person image and optional pose reference.Source for Reference images (opens in a new tab)
Output formats: WebP, JPEG, PNGSource for Output formats (opens in a new tab)
Usage metering: Native pricing depends on garment count and turbo mode; standard and turbo tariffs are alternatives.Source for Usage metering (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.

Pruna API

Provider source (opens in a new tab)

API model ID: p-image-try-on
Access: Hosted API; account and API key required. No generation was submitted.
Regions: Not established
API endpoint URL: POST https://api.pruna.ai/v1/predictionsSource for API endpoint URL (opens in a new tab)
Authentication: apikey and Model headers; request fields are nested under input.Source for Authentication (opens in a new tab)
Model controls: seed, prompt, turbo, person_image, output_format, garment_images, output_quality, reference_pose, preserve_input_sizeSource for Model controls (opens in a new tab)
Required request fields: person_image, garment_imagesSource for Required request fields (opens in a new tab)
Request quotas: 500 requests/minuteSource for Request quotas (opens in a new tab)
Asynchronous generation: Asynchronous by default. Try-Sync=true waits up to 60 seconds; a prediction ID and status URL support longer jobs.Source for Asynchronous generation (opens in a new tab)
Result retrieval: GET /v1/predictions/status/{id}; download the result from the returned URL.Source for Result retrieval (opens in a new tab)
Reference guidance: One garment per category. Duplicate categories, eyewear, gloves and handheld accessories are unsupported by this route. The provider refers eyewear to p-try-on-glasses, whose contract is not yet reviewed.Source for Reference guidance (opens in a new tab)
Image input limits: Required person_image and garment_images. Up to 11 garment references, up to 6 recommended. The schema describes the count but does not enforce maxItems.Source for Image input limits (opens in a new tab)
Generation controls: turbo defaults to false and is discouraged above 4 garments. preserve_input_size defaults to true. Optional reference_pose changes the pose before dressing.Source for Generation controls (opens in a new tab)
Prompt guidance: Optional prompt defaults to empty. Experimental text guidance selects garments from non-flatlay or multi-item reference photos.Source for Prompt guidance (opens in a new tab)
Image quality settings: output_quality 0–100, default 95; output_format defaults to jpg.Source for Image quality settings (opens in a new tab)
Pricing coverage: Product guide: standard $0.015 for first garment plus $0.008 per additional garment; turbo $0.008 per garment. API guide instead says 15/8 units per input image without explaining which inputs count.Source for Pricing coverage (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 15 Oct 2026
Category	Usage	Rate	Unit
Virtual try-on	Standard · first garment¹	$0.015	first garment
Virtual try-on	Standard · each additional garment²	$0.008	additional garment
Virtual try-on	Turbo · each garment³	$0.008	garment

Rate conditions

Standard mode. Add the additional-garment charge after the first. Person and pose-image surcharges are not established.
Standard mode, after the first garment. Do not add turbo pricing.
Turbo mode replaces standard first/additional pricing. Provider cautions against more than 4 garments.

Pricing source (opens in a new tab)

Data & enterprise terms

Pruna API terms retain customer ownership of input/output and permit processing to provide the service. Operational telemetry is used for security and service improvement. A universal retention period and zero-training guarantee are not established.

Read the service policy (opens in a new tab)

Replicate

Provider source (opens in a new tab)

API model ID: prunaai/p-image-try-on
Access: Hosted API; account and API key required. No generation was submitted.
Regions: Not established
API endpoint URL: POST https://api.replicate.com/v1/predictionsSource for API endpoint URL (opens in a new tab)
Authentication: Bearer API token. Supply the model/version and an input object.Source for Authentication (opens in a new tab)
Model revision: 863911421152e6ffffcd9c70192d93d6e56083721253347e8fad8fa67f971d72Source for Model revision (opens in a new tab)
Model selection: Version-pinned selector prunaai/p-image-try-on:863911421152e6ffffcd9c70192d93d6e56083721253347e8fad8fa67f971d72Source for Model selection (opens in a new tab)
Model controls: seed, no_op, turbo, prompt, person_image, output_format, garment_images, output_quality, reference_pose, preserve_input_sizeSource for Model controls (opens in a new tab)
Required request fields: person_image, garment_imagesSource for Required request fields (opens in a new tab)
Asynchronous generation: Prediction API; Prefer: wait can wait up to 60 seconds, with status polling for unfinished jobs.Source for Asynchronous generation (opens in a new tab)
Output formats: webp, jpg, pngSource for Output formats (opens in a new tab)
Result metadata: The model output schema is an image URI.Source for Result metadata (opens in a new tab)
Request quotas: Not collected
Image input limits: Person image and garment URL array required; up to 6 garment references recommended and 11 supported in the description. No maxItems constraint.Source for Image input limits (opens in a new tab)
Generation controls: turbo=false and preserve_input_size=true by default. reference_pose is experimental and can fail for some seeds.Source for Generation controls (opens in a new tab)
Image quality settings: JPEG default; quality 0–100, default 95.Source for Image quality settings (opens in a new tab)
Documentation discrepancy: README prices per garment, but current billing metadata publishes $0.007/run plus $0.008/input image megapixel. Rounding and which image inputs are metered are not specified; no exact request total is inferred.Source for Documentation discrepancy (opens in a new tab)
Billing basis: Published billing configuration: fixed run fee plus input image megapixels. This differs from native garment pricing.Source for Billing basis (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 15 Oct 2026
Category	Usage	Rate	Unit
Virtual try-on	Fixed run charge¹	$0.007	run
Virtual try-on	Input image usage²	$0.008	input image megapixel

Rate conditions

Add the input-megapixel charge. Conflicts with README garment pricing; actual metered input selection and rounding remain unverified.
Additional to the run charge. Which person, garment and pose inputs count, and any rounding or minimum, are not established.

Pricing source (opens in a new tab)

Data & enterprise terms

Retention, training use, regional processing and enterprise terms have not been summarized for this endpoint.

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.

Pruna API

p-image-try-on

Not verified

No verified model prompt-cache contract or caching discount. Uploaded assets and asynchronous jobs are not prompt caching.

How it works: Not collected
Minimum reusable input: Not collected
Lifetime & refresh: Not collected
How to enable it: Not collected
Usage fields: Not collected
Billing rules: Not collected
Limits & data boundaries: Not collected

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

Caching reviewed 15 Oct 2026

Replicate

prunaai/p-image-try-on

Not verified

No verified model prompt-cache contract or caching discount. Uploaded assets and asynchronous jobs are not prompt caching.

How it works: Not collected
Minimum reusable input: Not collected
Lifetime & refresh: Not collected
How to enable it: Not collected
Usage fields: Not collected
Billing rules: Not collected
Limits & data boundaries: Not collected

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

Caching reviewed 15 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

Named hosted image-processing component. The exact underlying checkpoint and private-weight provenance are not established; this is explicitly a managed service record, not an additional claimed private foundation checkpoint.

Classification evidence (opens in a new tab)

Sources reviewed

Pruna try-on garment categories, limits and mode pricing (opens in a new tab)Reviewed 15 Oct 2026

Pruna native OpenAPI 0.6.0 (opens in a new tab)Reviewed 15 Oct 2026

Pruna API terms, updated October 13, 2026 (opens in a new tab)Reviewed 15 Oct 2026

Pruna official provider branding (opens in a new tab)Reviewed 15 Oct 2026

Pruna virtual try-on product page and acknowledgments (opens in a new tab)Reviewed 15 Oct 2026

P-Image-Try-On on Replicate: model README (opens in a new tab)Reviewed 15 Oct 2026

Pruna virtual try-on native guide (opens in a new tab)Reviewed 15 Oct 2026

P-Image-Try-On Replicate schema and published billing configuration (opens in a new tab)Reviewed 15 Oct 2026

Model notes

Native P-Image and P-Video lineup, adapter training and selected Replicate/Ideogram access reviewed. Video workflows are explicitly managed components where private checkpoint identity is unresolved. Remaining host routes, reference/eyewear variants and historical details are under review.

Official Pruna provider icon; no separate version-specific artwork established.

Availability is based on public documentation. No paid inference or authenticated API probe was performed.

Request limits and prices remain deployment-specific; image charges are not converted into language-model token or question estimates.

Check against your workload

Try representative inputs, measure task success and total cost, then check the exact endpoint’s data policy.

Read our methodology