Deployment

Deploy Structure-D as a FastAPI HTTP service or use the high-performance Rust CLI.

FastAPI service

Structure-D ships a production-ready FastAPI application factory. Start it with Uvicorn:

bash 
# Development
uvicorn structure_d.api.app:create_app --factory --reload --port 8080

# Production
uvicorn structure_d.api.app:create_app \
  --factory \
  --host 0.0.0.0 \
  --port 8080 \
  --workers 4

Or programmatically:

python 
import uvicorn
from structure_d.api.app import create_app

app = create_app()

if __name__ == "__main__":
    uvicorn.run(app, host="0.0.0.0", port=8080)

API endpoints

All routes are mounted under /api/v1:

MethodPathDescription
GET/api/v1/healthHealth check
GET/api/v1/modelsList registered models
GET/api/v1/schemasList built-in schemas
GET/api/v1/formatsList supported file formats
POST/api/v1/extractUpload a file and extract structured data

Example extraction request:

bash 
curl -X POST http://localhost:8080/api/v1/extract \
  -F "file=@invoice.pdf" \
  -F "schema_name=key_value" \
  -F "task=extraction" \
  -F "save_format=jsonl"

Rust CLI

The CLI provides the same extraction pipeline as a native binary with no Python overhead:

bash 
# Extract from a single file
structure-d extract invoice.pdf --schema key_value --output output/

# Batch extraction from a directory
structure-d batch invoices/ --schema key_value --concurrent 8

# List available schemas / models / formats
structure-d schemas
structure-d models
structure-d formats

# Use a custom config
structure-d extract doc.pdf --config configs/production.yaml

Docker

Dockerfile dockerfile 
FROM python:3.12-slim

WORKDIR /app

# Install system deps for Tesseract OCR (optional)
RUN apt-get update && apt-get install -y tesseract-ocr && rm -rf /var/lib/apt/lists/*

COPY pyproject.toml .
RUN pip install --no-cache-dir ".[ingestion,api,llm]"

COPY . .

EXPOSE 8080
CMD ["uvicorn", "structure_d.api.app:create_app", "--factory", "--host", "0.0.0.0", "--port", "8080"]
bash 
docker build -t structure-d .
docker run -p 8080:8080 \
  -e OPENAI_API_KEY=sk-... \
  -e SD_INFERENCE__PROVIDER__PROVIDER=openai \
  structure-d

Environment variables

VariableDescription
OPENAI_API_KEYOpenAI API key
ANTHROPIC_API_KEYAnthropic API key
GOOGLE_API_KEYGemini API key
SD_INFERENCE__PROVIDER__PROVIDEROverride inference provider
SD_LOG_LEVELLog level: DEBUG / INFO / WARNING / ERROR
SD_API__PORTFastAPI server port (default: 8080)
SD_STORAGE__OUTPUT_DIROutput directory for saved results

API authentication is a Phase 3 feature

The current FastAPI service has no authentication layer. Do not expose it directly to the internet without a reverse proxy and authentication middleware.

Edit this page on GitHub