Tables

Use POST /v1/tables when table structure is the primary output, or when reviewers need table-cell evidence before building an Extract schema.

Request contract

{
  "input": {"file_id": "file_invoice"},
  "execution_mode": "sync",
  "grounding": "cell",
  "boundary_source": "detector",
  "detection_policy": "strict"
}

cURL

curl -sS -X POST "https://api.docspeed.ai/v1/tables" \
  -H "Authorization: Bearer ${DOCSPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "input": {"file_id": "file_invoice"},
    "execution_mode": "sync",
    "grounding": "cell",
    "boundary_source": "detector",
    "detection_policy": "strict"
  }'

Python

import requests

response = requests.post(
    "https://api.docspeed.ai/v1/tables",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json",
    },
    json={
        "input": {"file_id": "file_invoice"},
        "execution_mode": "sync",
        "grounding": "cell",
        "boundary_source": "detector",
        "detection_policy": "strict",
    },
    timeout=600,
)
response.raise_for_status()
print(response.json())

TypeScript

const response = await fetch("https://api.docspeed.ai/v1/tables", {
  method: "POST",
  headers: {
    Authorization: "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    input: { file_id: "file_invoice" },
    execution_mode: "sync",
    grounding: "cell",
    boundary_source: "detector",
    detection_policy: "strict",
  }),
});

console.log(await response.json());

Example response fragment

{
  "grounding": "cell",
  "tables": [
    {
      "html": "<table><tr><th>Description</th><th>Amount</th></tr><tr><td>Consulting</td><td>1240.00</td></tr></table>",
      "page_index": 0,
      "source_page_indexes": [0],
      "grid_row_count": 2,
      "grid_col_count": 2,
      "region_ids": ["p1_table_1"],
      "mapping_quality": "exact"
    }
  ],
  "grounding_regions": [
    {
      "region_id": "p1_table_1",
      "page_index": 0,
      "shape": {"type": "bbox", "x": 80, "y": 460, "w": 760, "h": 220}
    }
  ]
}

Production notes

• Use grounding: "cell" for AP, reconciliation, and exception-review workflows.
• Use boundary_source: "detector" unless you have a known layout boundary to reuse.
• Keep detection_policy: "strict" for production and switch to liberal only when evaluating recall on messy documents.