Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Photoroom / Photoroom image APIs

AI Backgrounds v2

Managed product-background generation with text instructions and image guidance on the Image Editing API.

Image generation
deprecatedClosed-source managed componentImage editing
Official documentation (opens in a new tab)Reviewed 15 Oct 2026
6 of 21 specification fields collected
Output formatsPNG, JPEG, WebP, AVIFSee specification notes below
Output formatsPNG, JPEG, WebP, AVIFSee specification notes below
Input billingSee ratesPhotoroom Image Editing 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

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Maximum input tokens
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Maximum output
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Maximum prompt characters
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Reasoning
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Tool calling
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Structured output
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Streaming
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Batch processing
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Fine-tuning
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Context caching
Not supportedSource for Context caching (opens in a new tab)

Identical Image Editing API calls are billed separately; no result caching available.

Knowledge cutoff
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Release date
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Parameters
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Architecture
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Languages
Not collected

Not established for this exact image model or managed deployment. No language-model capabilities are inferred from HTTP or JSON transport.

Input modalities
Text, ImageSource for Input modalities (opens in a new tab)
Output modalities
ImageSource for Output modalities (opens in a new tab)
Output formats
PNG, JPEG, WebP, AVIFSource for Output formats (opens in a new tab)
Image input limits
30 MB; resized to a 5,000-pixel longest edgeSource for Image input limits (opens in a new tab)
Negative prompt support
SupportedSource for Negative prompt support (opens in a new tab)

background.negativePrompt only affects selector 2; v3 and Studio ignore it.

Access & pricing

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

Photoroom Image Editing API

Provider source (opens in a new tab)
API model ID
2

Selected with pr-ai-background-model-version HTTP header. Not the internal research version ID.

Access
API key required
Regions
Eastern USA (provider API FAQ); finer deployment regions not specified
API endpoint URL
GET or POST https://image-api.photoroom.com/v2/editSource for API endpoint URL (opens in a new tab)
Authentication
x-api-key headerSource for Authentication (opens in a new tab)
Request quotas
Default 60 images/minute; larger Enterprise quotas by agreement. Parallel calls do not bypass the quota.Source for Request quotas (opens in a new tab)

May 2025 changelog says existing customers retain their prior settings.

Bulk processing
One image per call; parallel requests supported within account rate limits. No discounted asynchronous Batch contract established.Source for Bulk processing (opens in a new tab)
Request format
GET /v2/edit accepts imageUrl and query parameters. POST /v2/edit uses imageFile with multipart form fields; do not put POST options in the query string.Source for Request format (opens in a new tab)
Input file formats
PNG, JPEG, WebP, AVIFSource for Input file formats (opens in a new tab)
Output formats
PNG, JPEG, WebP, AVIFSource for Output formats (opens in a new tab)

PNG default; JPEG cannot preserve transparency. JPEG/WebP quality 90, AVIF quality 75 in the guide.

Input limits
30 MB file maximum. Larger images are resized to a 5,000-pixel longest edge. 25 megapixels is a recommendation, not an additional rejection threshold.Source for Input limits (opens in a new tab)
Output sizing
outputSize: auto (default), originalImage, croppedSubject or WIDTHxHEIGHT. Auto uses template dimensions when templateId is provided, otherwise original-image dimensions.Source for Output sizing (opens in a new tab)
Billing conditions
Plus plan: $0.10 per successful call, including combined edit options. Identical repeated requests are billed again. Enterprise/custom contracts can differ.Source for Billing conditions (opens in a new tab)
Free usage allowance
Watermarked sandbox: 1,000 calls/month, maximum 100/day. Prefix API key with sandbox_. No request was submitted during this review.Source for Free usage allowance (opens in a new tab)
Documentation discrepancy
Dedicated guide exposes background.expandPrompt.mode; captured OpenAPI exposes background.expandPrompt and lacks the selector header. Retain both descriptions as a schema/guide discrepancy. FAQ omits AVIF, which the dedicated format guide supports.Source for Documentation discrepancy (opens in a new tab)
Model selection
pr-ai-background-model-version: 2Source for Model selection (opens in a new tab)
Required request fields
Original image plus background.prompt, a guidance image, or both. background.imageFile/imageUrl is a static background, not guidance, and cannot be combined with background.prompt.Source for Required request fields (opens in a new tab)
Seed controls
background.seed accepts a positive integer; fixed seeds encourage similar results, not a byte-identical output guarantee. Exact upper bound not specified.Source for Seed controls (opens in a new tab)
Prompt expansion
Automatic expansion enabled by default. Guide uses background.expandPrompt.mode=ai.never to disable it.Source for Prompt expansion (opens in a new tab)
Reference guidance
background.guidance.imageFile or imageUrl; guidance.scale 0–1, default 0.6. Text and image guidance can be combined. Exact oldest-v2 guidance feature parity is not established.Source for Reference guidance (opens in a new tab)
Negative prompts
background.negativePrompt is effective.Source for Negative prompts (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 15 Oct 2026
CategoryUsageRateUnit
Image API usageImage editing, including generated backgrounds1$0.10successful API call

Rate conditions

  1. Published Basic/Plus plan rate. A single combined edit call is charged once, not once per enabled feature. Separate calls are billed separately; account-specific Enterprise pricing may differ. No text-token or per-question conversion.
Pricing source (opens in a new tab)

Data & enterprise terms

Provider FAQ says API images are not saved. This is not a full account-data, log-retention or training-use policy.

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.

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

Managed deployment, not a separately established private base checkpoint. Photoroom documents this model-backed workflow, but exact private derivative identity and the mapping to Instant Diffusion or third-party models remain unresolved. API versions, presets and transports do not establish new foundation models.

Classification evidence (opens in a new tab)

Sources reviewed