The creative layer for your product
Give your product
a creative layer.
Embed image, video, standalone audio, and text workflows inside the product your users already know. Saved context can steer each draft before your team reviews it.
✦ ImageLayer Sample
Create a launch campaign for our new botanical sparkling water.
Built for product teamsSaved brand contextNo framework lock-inAuthenticated API
02 — Native by design
Not another tab.
A new capability.
External AI tools break context, brand, and momentum. ImageLayer lives inside your product, carries its configured context and controls, and feels like part of the workflow.
Campaign / Spring launch
Saved just nowBuild the launch story
Your users stay in flow.
Your brand controls stay in view.
- 4
- media modes
- 1
- native workflow
- 0
- ML infrastructure
03 — One intent, many forms
Make a whole campaign.
Not just an asset.
Image, video, standalone audio, and text requests can share saved context. That context steers the model; it does not guarantee consistency between outputs.
ImageVideoVoiceText
01
Image
A fictional brand exploration
An illustrative set of campaign, social, product, and template drafts that still needs brand review.
02
Campaign
From plan to pipeline
Explore how a fictional product story can be translated into channel-specific drafts for review.
03
Commerce
Commerce concept drafts
Request catalog and campaign concepts using saved context, then verify product fidelity and claims.
04
Saved context
Visual, copy, and standalone audio
Use the same saved context for separate visual, text, and single-speaker audio drafts; review each output independently.
✦ Fictional product examples for interface exploration — not customer work or performance proof.
Explore product use casesInterface preview — no generation request
Inspect the workflow.
Configure the layer.
Switch between image, video, audio, and text to inspect the widget states. This public preview uses sample data and does not call a generation model.
Interactive UI preview · no account neededTo submit a request, sign in and review the enabled model and current quote.Open the playground
04 — Prepared integration surface
Connect the widget.
Keep authority server-side.
<image-layer id="creative-layer"
theme="auto" />
<script>
const layer = document.querySelector('#creative-layer')
layer.apiKey = 'il_pub_...'
layer.sessionToken = sessionToken
</script>- 01
Define the brand
Save visual direction, colors, tone, content types, and the generation modes eligible users may request.
- 02
Embed and authenticate
Place the native Web Component in the client and have your backend issue a short-lived, tenant-scoped session token.
- 03
Admit eligible requests
ImageLayer coordinates catalog checks, jobs, delivery, and usage after the workspace entitlement and spend controls pass.
- Read the documentation Open playground
ReactVueNext.jsAngularSvelteAstroPlain HTML
05 — Simple by default
Start with a layer.
Grow into a platform.
Create a workspace to inspect the catalog and configure the widget. The Free workspace includes no provider-backed generations; paid checkout is available only when the billing runtime is enabled.
Free
Configure a workspace and inspect the widget UI. No generation credits are included.
$0
Create Workspace- Workspace and widget configuration
- Public UI preview with sample data
- Up to 1 end user
- No included generation credits
Starter
For teams adding image, video, voice, and text workflows without building AI infrastructure.
$59/mo
Review Starter- 900 credits / month
- Up to 20 end users
- 10GB archive storage (30-day archive)
- Image generation
- Sketch inputs & Reference inputs
- Video generation (720p)
- Standalone single-speaker audio generation
- AI text generation
Pro
For growing teams that need richer media workflows, brand control, and usage visibility.
$199/mo
Review Pro- 3,000 credits / month
- Up to 50 end users
- 100GB archive storage (12-month archive)
- Image generation
- Sketch inputs & Reference inputs
- Video generation (1080p)
- Standalone single-speaker audio generation
- AI text generation
- Brand guidelines
- Usage analytics
Enterprise
For organizations that need scale, rollout support, and custom workflow control.
Custom
Talk to Sales- 10,000 credits / month
- Up to 100 end users
- 1,024GB archive storage (Custom archive)
- Image generation
- Sketch inputs & Reference inputs
- Video generation (1080p)
- Standalone single-speaker audio generation
- AI text generation
- Brand guidelines
- Usage analytics
- Custom integrations
Confirm each generation quote
Model availability and credit cost depend on the selected model, settings, workspace plan, provider eligibility, and current catalog.
The model selector and pre-submit quote are the authoritative sources for the request in front of you. Generation is admitted only after the current entitlement, credit balance, spend controls, and provider availability pass validation.
Plan prices and included paid-plan credits are shown from the shared application catalog. A visible paid tier does not mean checkout is enabled in the current runtime.
06 — Security & trust
The invisible layer
matters just as much.
Your data, your clients, your reputation. Built for products that can't afford a security incident.
Encrypted in transit
Connections use HTTPS/TLS. API keys are SHA-256 hashed and passwords are stored as bcrypt hashes.
Email-verified accounts
Organizations must verify their email before creating API keys or using live API flows.
Rate-limited & monitored
Built-in rate limiting with fail-closed behavior. Authentication events are logged for security analysis.
Short-lived generated media
Temporary generated media is cleaned up on a rolling TTL. Our current target is roughly 72 hours for generated assets.
IDOR protection
API requests are scoped to the authenticated organization, with signed-access patterns and tenant checks on stored assets.
Bot & abuse prevention
Cloudflare Turnstile on registration. Disposable emails are blocked. IP-based daily limits prevent mass sign-ups.
Want to learn more about our security practices?Read our security page
07 — Clear before you commit
Questions,
answered.
If your question isn't here, reach out and include the workspace, model, and workflow you are evaluating.
How does ImageLayer carry my brand context?
Two layers of context and control. First, you set a style description, color palette, tone keywords, and elements to avoid. Those instructions shape supported workflows toward your brand direction, but generated output still needs review. Second, image outputs can add logo and layout overlays in the branding editor after generation. Workspace and plan settings determine which controls your users can access.
What AI models power ImageLayer?
The current public catalog contains approved Google and OpenAI models: Gemini 3.1 Flash Image, Gemini 3 Pro Image, and GPT Image 2 for images; Veo 3.1 Fast and Quality for video; Gemini 3.5 Flash and GPT-5.4 mini for text; plus admitted Gemini and OpenAI models for standalone single-speaker audio. The exact catalog depends on workspace plan, provider configuration, region, account eligibility, and current availability. The public playground is a UI-only sample and does not call a model; authenticated generation still requires an admitted model and eligible billing entitlement.
What can ImageLayer generate?
The current launch path supports images, short video, standalone single-speaker audio, and text when the selected workflow is admitted for the workspace. Multi-speaker podcast audio and merged video voiceover are not enabled for production launch.
What are content type templates?
Pre-built forms organize inputs for common creative outputs such as stat highlights, quote cards, blog promos, announcements, and infographics. Users fill structured fields that become part of the request. The dashboard shows the current template limit for each workspace plan.
Does ImageLayer support social media image sizes?
Built-in presets request target dimensions and layout guidance for supported LinkedIn, Instagram, X, Facebook, Google Business, WhatsApp, Telegram, Pinterest, TikTok, YouTube, blog, and email formats. Model output and downstream platform crops can vary, so inspect dimensions, safe areas, text, and composition before publishing. Choose Custom for another supported aspect ratio.
How hard is the integration?
Add the native <image-layer> Web Component to the client, configure its public settings, and create short-lived tenant-scoped session tokens from your backend. The same authenticated boundary can expose only the image, video, standalone audio, and text workflows admitted for the workspace.
Does it work with my tech stack?
The widget is a native Web Component. ImageLayer documents integration patterns for React, Vue, Next.js, Angular, Svelte, Astro, and plain HTML. Every production embed also needs a host backend capable of issuing short-lived sessions; review the relevant guide for framework-specific loading and lifecycle details.
Can I control what my users generate?
Yes. You can control brand rules, content templates, enabled modes, and other dashboard settings for each workspace. Your dashboard also shows who created what, when, and how many credits were used.
What happens if I hit my credit limit?
A generation request fails closed when the workspace lacks enough eligible credits or another spend control rejects it. Checkout and billing-portal actions work only when the runtime billing mode is enabled. Credit packs are limited to eligible Starter and Pro workspaces; the Free workspace cannot use packs. Confirm the final quote and credited balance before submitting.
What video models does ImageLayer use?
The current public video catalog contains Google Veo 3.1 Fast and Veo 3.1 Quality. Supported launch settings are shown in the authenticated selector and can vary by plan, region, account eligibility, provider availability, duration, and resolution. The pre-submit quote is authoritative for the request.
How does the text-to-speech work?
The current launch path supports standalone single-speaker audio. Choose an admitted Gemini or OpenAI speech model and an available voice in the authenticated selector, submit a reviewed script, then listen to the generated file before publishing. Available models, voices, formats, limits, and quoted credits come from the current workspace catalog.
Can I generate podcast-style content with multiple speakers?
Not in the current production launch path. Multi-speaker podcast snippets fail closed while that workflow is being stabilized; the supported audio path is standalone single-speaker generation.
Is my data secure?
Connections use HTTPS/TLS. API key values are SHA-256 hashed before database storage, passwords are stored as bcrypt hashes, and temporary generated media is cleaned up on a rolling TTL. We require email verification before live API use and apply rate limits and tenant scoping. ImageLayer does not use your prompts or generated assets to train ImageLayer-owned models; configured AI providers process requests under their applicable terms. See our Security page for full details.
✦ Explore the integration surface
Configure the workflow.
Review the fit.
Create a workspace to inspect the catalog and integration settings. Generation is available only when the workspace has an admitted model and an eligible billing entitlement.