Inku
Inku is a Belgian platform that encodes a company's brand guidelines, assets and tone of voice into a queryable brand intelligence layer, then generates on-brand images, short video and campaign assets at scale.
What is Inku?
Inku is a Belgian platform that presents itself as an operating layer for brand-native marketing rather than as another image generator. Its premise is that a brand should stop living as a PDF in a shared drive and become a living, queryable model that people, tools and AI agents can all act on.
The product is built in four layers. The Brand Engine encodes guidelines, tone of voice, assets and product rules, including the unwritten conventions that usually leave with the next handover, into a brand intelligence layer that keeps growing as teams talk to it. It fine-tunes visual models, called Looks, on the customer's own identity, while a Taste Finder turns moodboards built from ten thousand royalty-free images into a usable model in one click. Inku claims more than ninety-nine per cent accuracy on product details such as materials, finishes, sizes and text.
Content Studio is the production layer: campaign briefings built inside the tool, reusable characters and worlds, images in 2K and 4K in every format, short HD video of five, eight or ten seconds, and advertising for digital, out-of-home, retail media and point of sale. Approval flows, comments, external sharing and an asset library with semantic search sit alongside. Newsletters, landing pages and scheduling are announced as coming.
AI Automation adds research agents that watch competitors, brand drift and cultural trends around the clock, draft creative briefs from what they find, and answer strategic questions with sourced reports. It connects natively to Meta MCP, and to Amazon, Shopify, Google and Snowflake.
Process Integration and Inku Labs cover the technical and human landing: an SDK exposing brand-native endpoints, an MCP server, process automation, deployment on Inku's cloud or inside the customer's own infrastructure, and forward-deployed marketers who work alongside client teams until adoption holds. The headline claims are ninety-nine per cent brand consistency, up to ninety per cent less lead time and up to eighty per cent budget savings, none of them supported by a published method. Named customers include Brussels Airlines, Duvel Moortgat, Loop Earplugs, DSTNY, Gimber, Just Russel and Collett & Victor. The product itself sits behind a login, with no self-serve sign-up on the public site.
What it does
- Encode brand guidelines, tone of voice, assets and product rules into a queryable brand intelligence layer
- Generate on-brand photography, pack shots and lifestyle imagery in 2K and 4K, in every format
- Train visual identity models, and convert a locked moodboard into a usable model in one click
- Produce short HD video of five, eight or ten seconds, plus ads for digital, out-of-home, retail media and point of sale
- Build campaign briefings with reusable characters and worlds, then decline them by region and by team
- Monitor competitors, brand drift and cultural trends around the clock and turn the signals into creative briefs
- Call the brand engine, briefing and generation layers from your own products through the Inku SDK or an MCP server
When to use Inku / When not to
A quick filter to help you decide if Inku is the right fit.
When to use Inku
- Brand managers at established consumer brands who need every asset to stay recognisably on-brand across markets
- Marketing teams producing high volumes of campaign variants for several regions, channels and languages
- Packaging and product-visual teams that need accurate pack shots and label detail without booking new photo shoots
- In-house creative directors who want internal teams and external agencies working from one shared brand layer
- Marketing operations and engineering teams ready to wire brand generation into their own tools through an SDK or MCP
When not to use Inku
- Individuals and small teams looking for a self-serve tool: there is no sign-up, no published price and no free plan
- Anyone who wants to evaluate the product alone, since onboarding runs through a pilot programme and the Inku Labs team
- Young brands with fewer than the twenty existing brand visuals the vendor gives as its minimum for good results
- Users who need a mobile app or a browser extension, as Inku is reachable only through its browser platform
- Buyers who require published terms of service, a data processing agreement or a named subprocessor list before signing
How to use Inku
A typical end-to-end flow, from setup to results.
- Start from the contact form on the company page, since there is no self-serve sign-up, or apply for the free two-hour workshop
- Attend the in-person "Getting Started" session in a group of four to six people, with live demonstration accounts provided
- Work through the six workshop steps: introduction, account access, brand building, refinement, briefings and campaigns
- Build the business case with Inku Labs, mapping current budget lines, channels and content needs against the tool's efficiency data
- Set up the brand engine through the pilot programme by supplying at least twenty existing brand visuals and product photos
- Let Inku train the models on the brand aesthetic, then review the generated examples and fine-tune until the output feels right
- Create campaign briefings inside the platform, locking reusable characters and worlds for teams and regions to work from
- Generate assets by chat or on the infinite canvas, adjusting aspect ratio, shot type and layout from a single panel
- Route work through the built-in approval and comment flows, then store final assets in the library for reuse
- Optionally connect data sources, wire the SDK or MCP server into internal tools, or deploy inside your own infrastructure
Pros & Cons
Pros
- Unusually explicit privacy posture for a visual generation tool: isolated environment per customer, dedicated models, no data shared between customers
- Infrastructure declared entirely European, publisher established in Belgium, and GDPR compliance claimed in plain terms
- Customers keep ownership of their data, their images and the trained models themselves
- Written commitment never to train general, shared models on customer content
- Seven named, detailed case studies with real European brands rather than anonymous testimonials
- Broad scope in one place, from brand encoding to production, approval, storage and automation
- Open technical route for engineering teams: SDK, MCP server and on-premise deployment for data residency constraints
Cons
- No public pricing at all: no pricing page, no rate card and no amount anywhere, so cost is only discoverable through a sales conversation
- No terms of service and no legal notice are published; the privacy policy is the site's only legal document
- The publishing entity is named nowhere on the site, the footer showing only "© 2026 Inku"
- No data processing agreement is published or offered, and no subprocessor is named
- No certification is attested for Inku itself; the site's only ISO mention concerns the open-source technology it uses
- The Trust page says data is never shared with external providers while the privacy policy describes third-party hosting, payment and analytics providers
- Performance figures of ninety-nine per cent, minus ninety per cent and minus eighty per cent are claimed without any published method or source
Pricing & Plans
Inku publishes no pricing. There is no pricing page, no rate card and no monetary amount anywhere on the site; the vendor states only that plans are flexible, tailored to enterprise needs, scale with the needs of the brand, and are agreed with its sales team. The cost structure is described as a licence plus Inku Labs service hours. No permanent free plan and no free trial of the product are announced. The only free item is the two-hour "Getting Started" workshop, held in person by application, with live demonstration accounts provided. The figures shown on the Inku Labs page, namely a 12.5 million EUR marketing budget, a 312,000 EUR investment and 2,890,000 EUR of value unlocked, are explicitly labelled an illustrative example and are not prices.
Data, GDPR & hosting
A consolidated view of how Inku handles your data.
GDPR overview
GDPR is addressed explicitly. The privacy policy, last updated in February 2026, lists the rights of access, rectification, erasure, restriction, portability, objection and withdrawal of consent, promises a response within thirty days through privacy@inku.tech, and recalls the right to lodge a complaint with a supervisory authority. The Trust page states that the entire infrastructure sits in Europe and presents this as ensuring GDPR compliance. Where international transfers occur, the policy relies on Standard Contractual Clauses approved by the European Commission. The service is not intended for anyone under sixteen. As the publisher is established in Belgium, no Article 27 representative is required and none is named. Two gaps remain: no data processing agreement is published or offered, and no subprocessor is named, the policy citing only categories such as hosting, payment and analytics providers.
Who owns the data?
Inku states that customers keep both what they bring and what they produce. The privacy policy, last updated in February 2026, says users retain all intellectual property rights over their brand assets and over generated content, and the Trust page extends that ownership to the trained models themselves. The declared controller is "Inku.tech", reachable at privacy@inku.tech; no legal entity is named anywhere on the site. Brand data is described as isolated per customer and inaccessible to other users. One caveat matters: no terms of service are published, so these commitments rest on a privacy policy and marketing pages rather than on a contract a buyer can read before engaging.
Reuse rights
Because Inku publishes no terms and conditions, reuse rights are set out only in the privacy policy and on the Trust page. Both state that the customer keeps full intellectual property rights over uploaded brand assets and over everything generated, and that the trained models belong to the customer as well, which points to unrestricted commercial reuse without asking permission. On Inku's own side, brand assets are processed to train AI models personalised to the account. The vendor commits in writing not to use customer content to train general models shared with other users, and describes each customer environment as fully isolated. No option to opt out of that personalised training is documented anywhere on the site.
Data retention & training
Hosting summary
Inku states that its entire infrastructure is located in Europe and repeats, in its FAQ, that everything runs on its own European servers, with all AI models trained on its own platform. Each customer is described as having a completely isolated environment whose data is never shared with other customers. No country, no region and no hosting provider is ever named, so the claim cannot be verified more precisely than "Europe". The privacy policy adds encryption in transit and at rest, access controls, regular security assessments and staff training, and provides for international transfers under Standard Contractual Clauses approved by the European Commission where they occur. For organisations with stricter requirements, Inku offers deployment inside the customer's own infrastructure, or a hybrid model with a central engine and local generation. Note that the public website is served through a content delivery network, whose node locations say nothing about where customer data is stored.
Things to keep in mind
Risks and trade-offs to weigh before adopting Inku.
- The commercial relationship cannot be assessed before contact: no terms of service, no legal notice and no named publisher on the site, so the only written document is the privacy policy
- No data processing agreement and no subprocessor list, which is a real obstacle for anyone about to entrust personal data or confidential brand material
- The Trust page claims data is never shared with external providers while the privacy policy describes third-party hosting, payment and analytics providers: read both before relying on either
- Delegating brand judgement to a model can quietly erode a team's own taste, so the approval flows are worth using as real editorial control rather than as a formality
- Generating unlimited variations makes it easy to flood channels with output nobody reads; volume is not the same as impact
- Headline figures of ninety-nine per cent consistency and eighty per cent savings come with no method, and the Inku Labs business case is explicitly an illustrative example, not a quotation
- The publisher is a very young company and the entire visual production chain would depend on it, so continuity and exit conditions deserve to be negotiated explicitly
Setup & Integrations
Technical difficulty
Low for users, moderate as a programme. No design skill and no prompt engineering are required: you describe what you want in plain language. Setting up is not self-serve, however. It runs through a pilot programme and the Inku Labs team, needs at least twenty existing brand visuals, and takes a few days from supplying material to the first assets. The vendor plans a running brand engine by month one or two and multi-team rollout by month three or four. Optional engineering work exists for teams wiring the SDK, the MCP server or an on-premise deployment.
Deployment
Integrations
Behind Inku
Resources
All the official URLs gathered for verification and reference.
Frequently asked questions
What is Inku?
How much does Inku cost?
Is there a free plan or a free trial?
Does Inku train AI models on my data?
Who owns the generated content and the trained models?
Where is the data hosted?
How much brand material do I need, and how long does onboarding take?
Does Inku offer an API?
Is there a mobile app, and is there a minimum age?
Should you pick Inku?
Inku is a more ambitious proposition than the phrase "AI image generator" suggests. The idea of encoding a brand once, as a living and queryable model, and then having briefings, production, approval, storage and monitoring all draw from that single source is coherent, and the four layers on the site fit together rather than merely coexisting. The data posture is the strongest part of the offer: an isolated environment per customer, dedicated models, a written refusal to train shared models on customer content, infrastructure declared entirely European, and ownership of data, images and trained models left with the customer. The seven named case studies, all real European brands with concrete before-and-after detail, give the claims more weight than testimonials usually carry.
The reservations are about transparency rather than capability. No price is published anywhere, no terms of service or legal notice exist on the site, and the publishing company is not named at all, which leaves a buyer with a privacy policy and marketing pages as the only written commitments before a sales conversation. No data processing agreement is offered and no subprocessor is listed, which procurement teams will notice. The Trust page and the privacy policy also disagree on whether external providers are involved. The performance figures are impressive and entirely unsourced, and a meaningful share of the roadmap, including newsletters, landing pages and scheduling, is still marked as coming.
Inku suits organisations with a strong brand, real volume, and the patience for a guided rollout with a young vendor. It does not suit anyone hoping to sign up, try it alone and compare prices in an afternoon.
- Choosing a selection results in a full page refresh.
- Opens in a new window.