The most accurate parser you can run yourself.

import base64, requests

pdf = base64.b64encode(open("annual-report.pdf", "rb").read()).decode()

r = requests.post(
    "http://localhost:5189/lmkit/v1/document-parsing",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={"input": pdf, "effort": "Medium", "output_format": "Markdown"},
)
print(r.json()["content"])

{
  "mcpServers": {
    "lmkit": {
      "url": "http://localhost:5189/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

using LMKit.Document.Parsing;

using var parser = new DocumentParser(ParsingEffort.Medium)
{
    IncludeFigureImages = true
};

ParsedDocument document = parser.Parse("annual-report.pdf");

File.WriteAllText("report.md",   document.ToMarkdown());
File.WriteAllText("report.json", document.ToJson());
document.SaveHtml("report.html");
document.SaveDocLangArchive("report.dclx");

using LMKit.Document.Parsing;

using var parser = new DocumentParser(ParsingEffort.High);
ParsedDocument document = parser.Parse("filing.pdf");

foreach (ParsedPage page in document.Pages)
foreach (ChartContent chart in page.Elements.Select(e => e.Content).OfType<ChartContent>())
{
    Console.WriteLine($"Page {page.PageNumber}: {chart.Caption}");
    Console.WriteLine(chart.Text);   // the plotted series as a table
}

using LMKit.Document.Parsing;

// On a connected machine: fetch the parser's model once.
DocumentParser.GetModelCards(ParsingEffort.High).Single().Download();

// On the air-gapped machine: point the parser at the file.
using var parser = new DocumentParser(ParsingEffort.High, @"models\lmkit-doc-parser-1.lmk");