Skip to content

Quickstart (server)

Run ReadAnyDoc as an HTTP server with Reducto's API, then call it with curl, any HTTP client, or Reducto's own SDK.

1. Start the server

pip install "readanydoc[engine]"
export ANTHROPIC_API_KEY=sk-ant-...
readanydoc serve --port 8000

git clone https://github.com/arpitpatel25/ReadAnyDoc && cd ReadAnyDoc
export ANTHROPIC_API_KEY=sk-ant-...
docker compose up --build

Check it is up:

curl http://localhost:8000/version
# "readanydoc-0.1.0"

2. Upload and parse

# upload a file, get a file id
curl -s -F "file=@invoice.pdf" http://localhost:8000/upload
# {"file_id": "reducto://5d44c8....pdf", "presigned_url": null}

# parse it
curl -s -X POST http://localhost:8000/parse \
  -H "content-type: application/json" \
  -d '{"input": "reducto://5d44c8....pdf"}'

The response has Reducto's shape: result.chunks[] with content and blocks[] (type, bbox, content, confidence), plus usage and job_id. A public URL works as input too.

3. Or use Reducto's SDK

from pathlib import Path
from reducto import Reducto          # pip install reductoai

client = Reducto(api_key="anything", base_url="http://localhost:8000")

upload = client.upload(file=Path("invoice.pdf"))
result = client.parse.run(input=upload.file_id)
for chunk in result.result.chunks:
    print(chunk.content)

The SDK requires an api_key string. It is only checked when you set READANYDOC_API_KEY on the server.

4. Before you share the server

  • Set READANYDOC_API_KEY so that every request needs Authorization: Bearer <key>.
  • Read Security and Configuration.

What's next