CutOptim Engine API

A deterministic cutting-optimization engine, callable over HTTP: the same request always returns the same cut plan — so you can audit it, quote from it, settle a dispute with it, and re-run last year’s order to get last year’s plan. It is the same engine that powers CutOptim, across the three rectangular modes (2D panels, 1D linear, and timber with cross-section matching), plus true-shape nesting for irregular polygon parts — POST /v1/optimize/nest, for laser, plasma and waterjet work. Send parts and stock, get back the full layout, cut plan and yield — ready to drop into an ERP, a quoting tool, or a machine’s own software.

See the full benchmark → · Measured, not claimed

Why build on it

What the API does

Which engine serves which mode

The endpoint you POST to picks the mode; the engine parameter picks the algorithm. The default heuristic engine serves 2D, 1D and timber; balanced and the asynchronous max engine are 2D only; and true-shape nesting runs on its own lbf engine. Same engine as the app, over HTTP.

• heuristic — The default — a guillotine multi-strategy packer. Highest yield, every layout is saw-cuttable, and it always returns a cut plan. It serves 2D, 1D and timber.
• balanced — An opt-in MaxRects free-nester for 2D. Much faster on very large jobs for slightly less yield, but its layouts often cannot be sawn edge-to-edge, so they carry no cut plan.
• max — An asynchronous tree search for 2D. It reaches the proven optimum on far more jobs, at seconds-to-a-minute per solve — you submit the job and poll GET /v1/jobs/{id} for the result. Still deterministic and saw-cuttable.
• lbf — The true-shape nesting engine. It packs irregular polygons into each other’s pockets for laser, plasma and waterjet; a higher-density sparrow build is planned.

Which endpoint for which material: flat sheets — plywood, MDF, glass, acrylic, sheet metal — go to /v1/optimize/2d; bars, tube, pipe and profile to /v1/optimize/1d; structural timber (a 50×150 only from 50×150) to /v1/optimize/wood; irregular polygons for laser, plasma and waterjet to /v1/optimize/nest.

One call, a full plan

curl https://api.cutoptim.com/v1/optimize/2d \
  -H "Authorization: Bearer co_live_…" \
  -H "Content-Type: application/json" \
  -d '{
    "parts": [
      { "name": "Door",    "w": 600, "h": 400, "qty": 4 },
      { "name": "Shelf",   "w": 800, "h": 300, "qty": 6 }
    ],
    "stock":   [{ "w": 2440, "h": 1220, "price": 42 }],
    "options": { "kerf": 3, "effort": "balanced" },
    "engine":  "heuristic"
  }'

Comes back as:

{
  "metrics": { "sheetCount": 1, "yieldPct": 80.62, "placed": 10, "total": 10,
               "cutLines": 10, "sawPasses": 13, "cutLength": 10720, "totalPrice": 42 },
  "sheets":  [ … ],
  "cutPlan": [{ "sheet": 0, "step": 1, "axis": "h", "pos": 400, "length": 2440, "stage": 1 }, … ],
  "guillotineValid": true,
  "engineVersion": "1.0.0+10e0c941",
  "deterministic": true
}

Full request and response shape, every option, all error codes and all three engines:API reference →

Fast or dense? Pick with one option

The effort option trades solve time against yield. Here is that trade, measured on one demanding job — every figure comes from the real packer.

And it is fast either way: even the largest production jobs — 2,000 parts and more — solve in single-digit seconds on the default engine, comfortably inside the API time budget.

Pricing

Larger volumes. 10,000 requests a month is the standard plan, not a ceiling on what we can run. If you need more — higher volume, separate keys for staging and production, or a dedicated arrangement — tell us your numbers and we price it individually. Tell us your volume →

The regular CutOptim plans (Free / Pro / Workshop) are unrelated to the API — the in-app optimizer stays included in them. See app pricing →

Solutions by industry

The same engine, positioned for the way one trade cuts. Each page shows the endpoint, a request and a response, and the fields that matter for that material.