API v1.4.2 [STABLE]
OpenAPI SpecPure Automation.
Seamlessly integrate industrial-grade AI pattern generation into your textile production pipeline. Our REST API handles the heavy compute, providing print-safe assets at scale.
Authentication
All API requests require a Bearer Token. Enter your key in the Authorization header.
Authorization: Bearer sk_live_51M_...Rate Limits
We utilize a dynamic token-bucket algorithm to ensure fair access. Enterprise tiers enjoy significantly higher burst capacities and dedicated compute lanes.
Standard Tier
60 RPM
Shared GPU Cluster
Enterprise Tier
1000+ RPM
Dedicated NVIDIA H100 Instance
Error Codes
| Code | Description |
|---|---|
| 401 | Invalid or missing API key. |
| 402 | Insufficient credits in account. |
| 429 | Rate limit exceeded. Slow down your requests. |
| 500 | Engine synchronization failure. Transactions are automatically refunded. |
Generate Pattern
POST /api/generate
Deploy the **Neural Core v4** specifically tuned for textile geometries. Direct prompt-to-tile support.
Parameters
- prompt: [string] Descriptive design query (min 3 chars).
- compositionType: [enum] "single" | "tile" (repeating).
- colorPalette: [array] List of HEX/HSL targets.
{
"prompt": "traditional damask, silk texture",
"aspectRatio": "1:1",
"density": "high",
"compositionType": "tile",
"output_format": "png"
}Vectorize
POST /api/vectorize
Industrial raster-to-SVG conversion. Zero-stroke commitment for cleaner CAM/CAD parsing.
curl -X POST /api/vectorize \ -H "Authorization: Bearer sk_..." \ -F "[email protected]" \ -F "quality=high" \ -F "config={"iterMax":20}"
Upscale 16x
POST /api/upscale
Response Object
{
"url": "https://cdn.sapphire.com/res_..."
"width": 16384,
"height": 16384,
"processingTimeMs": 12450
}Need dedicated compute?
Large-scale manufacturers can request dedicated GPU instances for guaranteed latency and private security models.
Contact Engineering