SoMark API

Best for teams that want direct HTTP control over request parameters, backend flow, and result handling. If you are still comparing integration paths, start with the get started overview; if you need auth, limits, and endpoint coverage, also check the API overview.

Get your API key and run the preflight check

Visit somark.ai/Studio/apikey to get an API key in the sk-*** format.Before calling the API, check three things:

api_key is included
the file is not empty and is supported
output_formats only uses markdown, json, or zip

See Free tier rules for quota details.

Complete one sync parsing request first

Start with POST /parse/sync for small files. The result comes back in the same response.

curl -X POST https://somark.ai/api/v1/parse/sync \
  -F "file=@document.pdf" \
  -F "output_formats=markdown" \
  -F "output_formats=json" \
  -F "api_key=sk-your-api-key"

import requests

with open("document.pdf", "rb") as file_obj:
    response = requests.post(
        "https://somark.ai/api/v1/parse/sync",
        files={"file": file_obj},
        data=[
            ("output_formats", "markdown"),
            ("output_formats", "json"),
            ("api_key", "sk-your-api-key"),
        ],
        timeout=120,
    )

payload = response.json()
print(payload["data"]["result"]["outputs"]["markdown"])

Use async submit and check for large files

Large files, batch jobs, and background workflows should use async mode: call POST /parse/async first, then poll POST /parse/async_check with the returned task_id.

curl -X POST https://somark.ai/api/v1/parse/async \
  -F "file=@large.pdf" \
  -F "output_formats=markdown" \
  -F "api_key=sk-your-api-key"

curl -X POST https://somark.ai/api/v1/parse/async_check \
  -F "task_id=your-task-id" \
  -F "api_key=sk-your-api-key"

Keep polling intervals around 3 to 5 seconds.

Open the detailed endpoint docs when needed

Getting Started Overview SoMark CLI

Overview

Quickstart

Ecosystem

FAQs