Source: https://seerepo.com/compare/datalab-to/marker/vs/khoj-ai/khoj
Markdown source: https://seerepo.com/compare/datalab-to/marker/vs/khoj-ai/khoj.md
Description: marker and khoj share 9 technologies: OpenAI, Google, Anthropic, transformers, torch, pillow, markdownify, Pydantic, and psutil. Both are written in Python.

# marker vs khoj

marker and khoj share 9 technologies: OpenAI, Google, Anthropic, transformers, torch, pillow, markdownify, Pydantic, and psutil. Both are written in Python.

- [datalab-to/marker](https://seerepo.com/r/datalab-to/marker): Convert PDF to markdown + JSON quickly with high accuracy
- [khoj-ai/khoj](https://seerepo.com/r/khoj-ai/khoj): Your AI second brain. Self-hostable. Get answers from the web or your docs. Build custom agents, schedule automations, do deep research. Turn any online or local LLM into your personal, autonomous AI (gpt, claude, gemini, llama, qwen, mistral). Get started - free.

## What they share and what sets them apart

- In common: OpenAI, Google, Anthropic, transformers, torch, pillow, markdownify, Pydantic, psutil
- Only in marker: surya-ocr, scikit-learn, markdown2, filetype
- Only in khoj: Django, FastAPI, pgvector, psycopg2-binary, Postgres, PostHog, Resend, anyio, AIOHTTP, uvicorn, electron-store, Electron, React, Next.js, Rich, itsdangerous, authlib, einops, openai-whisper, sentence-transformers, LangChain, e2b-code-interpreter, langchain-text-splitters, MCP, tiktoken, defusedxml, docx2txt, lxml, markdown-it-py, pymupdf, beautifulsoup4, magika, jinja2, email-validator, isomorphic-dompurify, Tailwind CSS, class-variance-authority, tailwind-merge, Radix UI, React Hook Form, Framer, cmdk, embla-carousel-autoplay, embla-carousel-react, Excalidraw, input-otp, intl-tel-input, Mermaid, vaul, @phosphor-icons/react, Lucide, katex, markdown-it, dompurify, @hookform/resolvers, Zod

## Side by side

| | marker | khoj |
|---|---|---|
| Size | 16,474 lines written, 121 code files | 71,518 lines written, 205 code files |
| Who writes it | 1% of its commits signed by AI. With Claude Code | 0% of its commits signed by AI |
| Activity | Last push October 29, 2026, 1,097 commits, 40,791 stars | Last push August 2, 2026, 4,763 commits, 38,022 stars |
| Tests | 36 test files. 37.2% of its code files have a test that imports them | 30 test files. 14.1% of its code files have a test that imports them |

- Size: Lines written by people and agents, without generated ones; code files, without tests.
- Who writes it: A commit counts as AI's when it carries its signature: a Co-Authored-By line, its author name or the merge of its branch.
- Activity: From GitHub and its history, on the day SeeRepo read it.
- Tests: The files that test the code, and the share of code files a test imports.

## Frequently asked questions

### Which is bigger, marker or khoj?

khoj, with 71,518 lines written; marker has 16,474.

The proof: [khoj-ai/khoj's code at commit ae229ca](https://github.com/khoj-ai/khoj/tree/ae229ca894c0b80ad84664afcfdde523b5e87057)

### Which has more GitHub stars, marker or khoj?

marker, with 40,791 stars; khoj has 38,022.

The proof: [datalab-to/marker on GitHub](https://github.com/datalab-to/marker)

### Which has more tests, marker or khoj?

marker, with 36 test files; khoj has 30.

The proof: [datalab-to/marker's code at commit 71dd5fd](https://github.com/datalab-to/marker/tree/71dd5fd7e51ab24f8624b6ba2fcb6014b4e40ed4)

### Which was updated more recently, marker or khoj?

marker: its last push was on October 29, 2026, and khoj's on August 2, 2026.

The proof: [datalab-to/marker's commits on GitHub](https://github.com/datalab-to/marker/commits)

### Which has more commits signed by AI, marker or khoj?

marker: an AI signed 1% of its commits, with Claude Code.

## Sources

- datalab-to/marker: Read on November 6, 2026, at commit 71dd5fd (https://github.com/datalab-to/marker/tree/71dd5fd7e51ab24f8624b6ba2fcb6014b4e40ed4)
- khoj-ai/khoj: Read on November 6, 2026, at commit ae229ca (https://github.com/khoj-ai/khoj/tree/ae229ca894c0b80ad84664afcfdde523b5e87057)

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.