LiquiFire Chains™
One chain language.
Every output.
A single declarative pipeline for imaging, 3D, data, and AI — authored once, stored anywhere, invoked by any system over HTTP. Teams ship visual experiences faster, maintain less, and build on work that compounds rather than duplicates.
Your visual operation is either a cost center or a compounding asset. LiquiFire Chains make it the latter.
Most organizations manage visual content through a collection of disconnected tools, each owned by a different team, each requiring its own expertise, its own contract, and its own maintenance budget. Every new capability requires a new integration. Every specialist who leaves takes institutional knowledge with them. LiquiFire Chains replace that accumulation with a single shared language — and every chain your organization writes permanently reduces the cost of every visual operation that follows.
Assess your current visual stackVisual variants that previously required a production sprint are authored as a chain parameter change. Campaign imagery, personalized content, and configured product visuals are generated at request time — no handoff to engineering, no queue, no delay.
LiquiFire Chains integrate over HTTPS, JSON, and XML. No new infrastructure is required. Existing DAM, ERP, PIM, and CDN investments are extended rather than replaced. The platform deploys as SaaS, managed cloud, or on-premise appliance — in the environment your security team approves.
Five or more vendor relationships with independent renewal cycles, usage-based pricing, and integration overhead consolidate into one platform agreement. The operational cost of maintaining fragmented tooling — documented at 20–40% of technology estate value — goes directly to new capability.
A declarative pipeline for everything visual.
A LiquiFire Chain is a sequence of named commands and arguments. Each command is plain English. The same syntax that scales an image also calls an AI model, queries a data table, or drives a 3D render. No proprietary SDKs. No framework dependencies. Any system that can construct an HTTP request can invoke a chain.
Chains execute sequentially and support conditional logic, loops, and nesting. Assets — images, fonts, color profiles, data tables, even other chains — are sourced from any valid URL at render time. Simple enough to write in a browser address bar. Powerful enough to run a production configurator serving millions of requests.
scale=size[800]
sink=format[webp],quality[90]
Both representations execute identically. No compilation. No deployment. The chain is the request.
Stored anywhere. Invoked by anything.
LiquiFire Chains run in two modes. The distinction shapes how teams work, how complexity is managed, and how quickly a new integration goes live.
Direct — the URL is the program.
Every command is passed inline as a query parameter. No file to manage, no deployment step. Any system that can construct an HTTP request can invoke a chain immediately. Useful for rapid prototyping, one-off operations, and integrations where the calling system controls all logic.
Stored — the URL passes values. The file holds the logic.
Production logic is encapsulated in a stored chain file and called by reference. The caller passes only the variables that change per request — product, color, size. The chain resolves everything else. Stored chains can be hosted on any accessible URL: resident storage, S3, a CDN edge, or an internal file server. Where the chain lives is your decision.
One language drives four domains.
Most organizations treat imaging, 3D, data processing, and AI as separate problems requiring separate tools. LiquiFire Chains handle all four from the same pipeline.
Photorealistic output from any source.
Source raster, vector, PDF, PSD, embroidery, and 50+ other formats. Apply colorization, compositing, typography, and format conversion in a single chain. Output to WebP, JPEG, PNG, TIFF, GIF, PDF — device-aware, CDN-ready, sized for every breakpoint.
3D renders invoked like any other command.
LiquiFire Chains drive 3D render operations, 360-degree output, and AR-ready asset generation through the same command model used for flat imaging. The same variable binding that populates a 2D swatch configurator populates a 3D scene.
Live data resolves at render time.
Chains read from lookup tables, XML, JSON, EXIF metadata, and remote data sources mid-execution. A RAL color code resolves to a hex value before the colorization command runs. A product ID resolves to a SKU description before text is annotated onto the image.
AI models as chain commands, switchable conditionally.
LiquiFire Chains call AI operations as first-class commands. Multiple AI models can be mixed within a single chain. Which model executes is determined by a conditional expression. Switching from one provider to another is a chain edit, not a code deployment.
Real chain. Real configurator. No modifications.
An unmodified production chain running a vehicle color configurator for an automotive OEM. Each block is annotated.
lookup=key[global.trailer],table[file:RAL-colors.table],name[trailercw]
lookup=key[global.chassis],table[file:RAL-colors.table],name[chassiscw]
blank=color[('global.trailercw.1')],name[cw1],if[(! global.storedasset)]
composite=compose[colorizeone],image[cw1],if[(! global.storedasset)]
composite=compose[multiply],image[trailermulti]
scale=size[global.outsize],if[('global.outsize' ne '')]
What consolidation actually delivers.
A unified rendering language changes more than the technology stack. It changes how fast teams move, how much they spend, and what they can build next.
Faster time to market
Teams that would spend weeks integrating a new imaging service and training against its SDK instead write a chain. New visual capabilities are typically live within days. When requirements change, the update is a chain edit, not a coordinated cross-team deployment.
Days vs. weeks to new capabilityLicensing consolidation
A single platform agreement replaces subscriptions for imaging libraries, AI service APIs, format conversion tools, and 3D rendering middleware. Vendor negotiation surfaces shrink from five or more renewal cycles to one. Usage-based AI billing becomes predictable inside a single commercial relationship.
One agreement, one renewal cycleFaster knowledge transfer
A rendering pipeline built across three languages and four vendor SDKs requires developers who understand all of them. LiquiFire Chains use one human-readable syntax across every domain. A developer joining the team reads existing chains on day one and contributes without a multi-week ramp. Institutional rendering knowledge is in the chain library, not in individual heads.
One syntax to learn across all operationsFreed engineering capacity
McKinsey CIO research documents that actively managing and consolidating technical debt frees up to 50% more engineer time for work that supports business goals. Teams that replace fragmented rendering stacks with LiquiFire Chains stop paying the maintenance tax and redirect that capacity to new product development.
Up to 50% more time on new developmentNot Python. Not Node. Not a bespoke SDK per vendor.
A single chain language with a consistent command syntax that developers and technical teams author directly.
Organizations managing visual content at scale accumulate dependencies: image processing libraries, AI service clients, 3D rendering plugins, format conversion tools. Each requires different expertise; each is a single point of failure when its specialist leaves. LiquiFire Chains replace that stack with one declarative language. A developer who knows chain syntax can author imaging operations, invoke AI models, process data, and trigger 3D renders without switching contexts or writing integration glue. The syntax is human-readable English. A non-developer can read a chain and understand what it does.
Work compounds. It does not duplicate.
Every chain ever written is a potential building block for every chain written after it. This is the architectural consequence of a shared language.
Build once. Call everywhere.
A chain authored for one project is callable from any other chain in any other project.
Color normalization, logo rendering, GeoIP resolution: build it once, store it, and call it by URL. Any chain on any system that can reach that URL can use it. No package registry, no version lock, no import statement.
The same chain. Every possible output.
Chains accept live variables at render time from any system that can construct an HTTP request.
PIM, ERP, CPQ, and commerce platforms pass variables as URL parameters or POST body. A single stored configurator chain produces a different output for every unique combination of color, material, size, and locale passed in.
Chains decide. You define the conditions.
Conditional execution, loops, and error branching are first-class chain features.
A chain can test whether a cached render exists before re-rendering, switch AI models based on content type, loop over a data set to generate multiple outputs, or take an alternate render path when a source asset is unavailable.
Self-contained visual objects inside the chain.
LiquiFire Ornaments extend the chain language with a concept that has no equivalent in conventional rendering pipelines: a self-contained visual unit that encapsulates its own rendering behavior, positional logic, attribute schema, and measurement units. An ornament is not a static asset. It is a named object that knows what it looks like, how it behaves, and can be placed, configured, and replicated inside any chain with a single command.
Ornaments are unit-aware: an attribute specified in inches renders correctly whether the output is screen resolution or print-ready TIFF. Schema validation runs before rendering, so if an attribute exceeds its declared maximum, the chain returns a structured error rather than a corrupt output.
source=url[file:door-base.png]
# Place a door handle ornament — position, finish, and hardware inside
ornament=id[door],scale[1.0],
attributes[{"finish": global.finish, "style": global.style, "position": "right"}]
sink=format[webp],quality[90]
Change one attribute. The ornament adapts.
width: 36in
hinge: left
glazing: yes
width: 24in
hinge: right
glazing: no
width: 36in
hinge: right
finish: oak
width: 24in
hinge: left
finish: black
A library that grows with every project.
Past work permanently reduces the cost of future work. LiquiFire ships a library of pre-built Foundation Chains covering the operations every enterprise imaging deployment eventually needs. Teams call them rather than rebuilding them.
Every chain an organization writes becomes part of its own chain library. A retailer's color normalization chain, called by their first configurator, is automatically available to every configurator, content pipeline, and print workflow they build afterward. The investment in one chain reduces the cost of every future chain that calls it.
Any system. Any source. Any environment.
LiquiFire Chains integrate over HTTP. Any system that can construct a URL or POST request already speaks the language — and deployment options range from SaaS to fully on-premise appliance.
No SDK required.
Every chain is invocable via a standard GET or POST request from any language, any platform, any integration layer.
Python, Node, PHP, Java, a shell script, a CDN edge function — if the system can construct an HTTP request, it can invoke a LiquiFire Chain. The rendering logic lives in the chain; the calling system passes variables and receives output.
Source assets from any URL.
Assets — images, fonts, data tables, color profiles, other chains — are sourced from any valid URL at render time.
HTTP, HTTPS, FTP, S3, and connected third-party protocols are all supported. Assets stay where they are. There is no asset migration, no format normalization step, and no staging environment required.
Rendered output cached at the edge.
Chains control cache behavior per request, balancing delivery speed against output freshness.
Static outputs cache at the CDN layer and serve at edge speed. Dynamic or personalized outputs bypass the cache on request. The chain decides, using nocache[false] for stable renders and nocache[true] when inputs are user-specific.
Your infrastructure, your rules.
LiquiFire Chains run in the environment your security and architecture teams require. Deployment model is a configuration choice, not a product constraint.
Managed cloud
Fully managed by LiquidPixels. No infrastructure to provision. Scales automatically. Available in multiple regions. Suitable for organizations that want full capability without operational overhead.
Dedicated, isolated tenancy
A dedicated LiquiFire environment provisioned in a private cloud tenancy. Data and rendered assets remain within your designated infrastructure boundary. Suitable for organizations with data residency requirements.
Runs inside your network
LiquiFire OS deployed as a physical or virtual appliance within your own network perimeter. No external data egress. Assets, source files, and rendered outputs never leave the secured environment. Suitable for regulated industries, financial services, and manufacturing with strict data governance requirements.
Split by workload or sensitivity
Combine deployment models by workload type. Public-facing rendering runs in SaaS. Sensitive product data or regulated workflows run on-premise. The same chain language operates identically across both — no rewriting, no retraining.
LiquidPixels has 25 years of enterprise deployment experience across manufacturing, luxury, retail, and regulated industries. We work with your architecture and security teams directly during evaluation.
The cost of the gap. The value of closing it.
Fragmented rendering stacks carry costs that compound over time. LiquiFire Chains close them with a single platform — and the freed capacity goes directly to new product development.
of the entire technology estate value consumed by technical debt, per McKinsey — before accounting for the new development it displaces.
McKinsey 2022of every engineer's working week spent managing existing fragmentation rather than building new capability, per Stripe research.
Stripemore engineer time available for new product work when technical debt is actively consolidated. The payoff is immediate and measurable.
McKinsey CIO researchOne platform agreement replaces many.
Each imaging library, AI service, and 3D rendering SDK carries its own commercial terms, renewal cycles, and usage-based pricing. Organizations managing five or more specialized tools regularly underestimate their true software spend by a factor of two to four, once integration, compliance, and operational overhead are included. LiquiFire Chains consolidate those agreements into one.
Rendering knowledge stays in the organization.
When expertise is distributed across three languages and four vendor SDKs, each departure creates a gap that is expensive to fill. With LiquiFire Chains, institutional rendering knowledge lives in the chain library — not in individual heads. Any developer who learns chain syntax can read, modify, and extend any chain ever written, regardless of who authored it or when.
One execution model means fewer seams where debt accumulates.
Fragmented stacks accumulate integration debt at every boundary between tools. LiquiFire Chains replace those boundaries with a single execution model. When a new AI model is available, the update is a chain edit. When a new output format is required, it is a sink parameter. Dependencies are managed in one place, and changes are localized to the chain that owns them.
One infrastructure. Every system. Full control.
LiquiFire OS is the core visualization infrastructure that connects to anything — Cloudinary, imgix, Adobe DAM, ERP, PIM, CPQ, and beyond. Rather than replacing what already works, the platform integrates across the entire technology estate and extends the capability of every system it touches. Data flows in from any source. Visual output flows out to any channel. The silos between teams, tools, and workflows dissolve.
Organizations running LiquiFire OS gain centralized control over their entire visual operation without disrupting existing infrastructure investments. Every integration adds capability in both directions — and every chain written compounds that value across every system connected to it.
Empower your organization with LiquiFire Chains™
We work with engineering and architecture teams to evaluate chain-based workflows against existing tooling — showing exactly where LiquiFire Chains reduce cost, compress timelines, and compound value across the organization.
Talk to an Expert