On-device engine · now onboarding design partners

Media intelligence for every app,
running on the device.

Embed transcription, semantic search, clip production and NLE interchange as a signed binary — scoped by a Grant, licensed on flexible terms. Your users' media never leaves their machine.

Signed binary · localhost /api/v1 · no cloud dependency

AvailableClip Production engine — turn any moment into a finished 9:16 clip, rendered on-device. No render-farm bill.
Select partnersMedia Understanding — transcription + semantic search across words, faces, OCR, logos, scenes and music. Indexed locally.
ModelThe Grant — one signed binary, projected into per-tenant capability. License one component without shipping the rest.
On requestNLE Interchange — clean FCPXML, EDL and OTIO with drop-frame timecode. Handoff to Final Cut, Premiere, Resolve.
On requestProvenance & Compliance — on-device attestation and integrity hashing. For legal, evidence and enterprise.

The pipeline

From raw file to finished moment — all on-device.

The engine ingests media, understands it, and produces from it, without a single byte leaving the machine. Embed the whole path, or just the stage you're missing.

Raw media

Video, audio, images — the files your users already have.

Index on-device

Transcribe, faces, OCR, logos, actions, scenes, music.

Semantic search

Find the exact moment by meaning, with an explainable trace.

Produce & export

Render a clip, or emit FCPXML / EDL / OTIO.

Your app

Results key back by Moment ID with file handoff.

Every stage runs behind localhost /api/v1 — nothing uploads, and the network can be off.

Components

A spectrum of components for your product.

The engine is modular. License the capability that fills your gap — enabled from the same binary by your Grant — and add more as you grow.

● Available

Clip Production

Turn any moment into a finished vertical clip inside your app — auto-reframe to 9:16, burned-in captions, highlight selection and render. No cloud round-trip.

For creator and clip tools, social schedulers, CapCut-style editors.

Get started

● Select partners

Media Understanding

Transcription plus semantic search across everything in a file — words, faces, OCR, logos, actions, scenes and music — indexed on-device with an explainable retrieval trace.

For DAM/MAM, podcast and meeting tools, archives.

Contact us

● On request

Interchange & Provenance

Clean FCPXML, EDL and OTIO with drop-frame timecode for real timeline handoff — plus on-device attestation, integrity hashing and a deterministic tool ledger.

For editor vendors, post tools, legal, evidence and enterprise.

Contact sales

Why on-device

Built to embed, not to operate.

The compute is already sitting on your users' machines. Embedding the engine turns that into your feature — with properties a cloud API structurally can't offer.

Nothing uploads

Indexing, search and rendering all run locally — "we never see your media" becomes an architecture claim, not a promise.

No inference bill

Compute runs on the device you don't pay for. Your gross margin stops tracking users' media volume.

Scoped by Grant

A signed key enables exactly the components you license and denies the rest — from one identical binary.

Flexible licensing

A flat per-integration license by default — or usage-metered, per-seat, or revenue share. Priced to your model.

Multimodal breadth

Words, faces, on-screen text, logos, actions, scenes and music — one engine, indexed together.

Nothing to operate

It ships as a signed binary. No model serving, no GPU fleet, no eval pipeline to staff.

Integration

Deploy anywhere. Call it over HTTP.

The engine runs as a local service on the platforms your product already targets. You ship the binary and your Grant; you call /api/v1 from your app.

macOS · Apple Silicon (signed) Windows Linux Desktop or server Air-gapped-capable
  1. 1

    Embed the signed binary

    Bundle the engine with your app. It runs as a local service exposing /api/v1 on loopback — no external host, no account for your users.

  2. 2

    Boot with your Grant

    The signed Grant scopes the capability subgraph. A clip-production Grant lights up create.* and leaves search retrieval denied.

  3. 3

    Call it from your app

    Drive it over HTTP. Results key back by Moment ID, with file handoff for rendered artifacts — so it composes with your data model.

  4. 4

    Report & bill

    The engine counts calls locally and reports usage — under metered terms you pay for what it does, or it's simply visibility under a flat or per-seat license.

Licensing

Flexible terms, priced to fit the deal.

Because the compute runs on your users' devices, we're not passing a GPU bill through to you — so we can shape licensing around your model. Most integrations start with a flat per-integration license; usage, per-seat and revenue-share terms are on the table too.

DefaultFlat

Per-integration license

One predictable fee for the component, regardless of volume — the simplest way to start a first integration.

Usage-metered

Pay per use

Per clip rendered or file indexed. Cost tracks the value the engine delivers.

Per-seat

Mirror your pricing

Aligns with your own model if your product already bills by seat.

Revenue share

Grow together

Lower fixed cost, shared upside — we're invested in the deal working.

Proven in production

Not a demo — a shipping engine.

Faring is the engine already powering MediaFind, a released on-device media search & production app, and a family of sibling products built on the same core. We license that proven pipeline to product teams — we're onboarding the first integration partners now.

Creator & clip tools

Add finished-clip production without a render backend.

DAM / MAM & archives

Make stored media searchable by what's inside it.

Editors & post tools

Hand moments off to a real NLE timeline.

Legal & enterprise

Prove processing ran on-device, with an audit trail.

Honest scope

What Faring is not.

We're deliberate about what we are — and clear about what we're not — so the fit is obvious before you invest engineering time.

  • Not a hosted cloud API — there's no Faring server your users' media flows through
  • Not a consumer app — that's MediaFind; Faring licenses the engine underneath it
  • Not a menu you buy whole — you license scoped components, starting with one
  • Not a foundation-model vendor — we ship an integrated media pipeline, not raw weights
  • Not generally available yet — we're onboarding design partners by hand

FAQ

Questions integration teams ask.

What exactly ships to my product?
A signed engine binary that runs as a local service exposing /api/v1 on loopback, plus a signed Grant key that scopes which capabilities are enabled. You call it over HTTP from your app. No cloud dependency, no model-serving infrastructure to run.
Does any user media leave the device?
No. Indexing, transcription, search and clip production all run on the end user's machine. The only expected network activity is license validation and usage metering — never the media itself.
Can I license just one capability?
Yes — that's the point of the Grant. It enables exactly the component you license (say, clip production) and denies the rest at the entitlement chokepoint, from the same binary. You can add components later without a new integration.
What platforms does the engine run on?
macOS on Apple Silicon is the primary, signed target today; Windows and Linux engine binaries are available for server and desktop integrations. Tell us your deployment target and we'll confirm the current state — we won't overstate what's ready.
How is it priced?
Flexibly — with a flat per-integration license as the default. Because compute runs on the user's device, we're not passing a GPU bill through to you, so we can fit the terms to your model: a flat license to start, or usage-metered (per clip or per file), per-seat, or revenue share. Design-partner terms are available while we onboard the first integrations.
Will you license the component that competes with me?
We're deliberate about competitive scope. The Grant lets us enable, for example, clip production for a tool that already has search, without handing over search retrieval. If a component is your core product, we'll say so plainly rather than arm a competitor.
How does it fit my data model?
Results key back to your app by Moment ID, with file handoff for rendered artifacts. You keep your own IDs, storage and UI; the engine is a capability behind your product, not a system of record.

Embed the engine. Keep the media on-device.

We're onboarding a small number of design partners for the first integrations. Tell us the gap you're filling and we'll scope a Grant for it.