Ka pūrongo te hapa / Te tono āhuatanga
PDF Extractor PDF to Text Academic Paper Extractor Research Paper to Text arXiv PDF Extractor More →

PDF to Markdown

Whakama ā-pūnaha OK 380+ tauira Kāore he tohu wai Kāore he kōwhiringa e hiahiatia ana
Kāhua:
+ GPT-5, Claude, Gemini
Drop a PDF — AI converts it into clean GitHub-flavored Markdown with headings, paragraphs, lists, tables, and code blocks all preserved. Powered by IBM Granite-Docling-258M (Apache 2.0). Faster + smarter than plain text extraction.

Drop a PDF here or click to upload

PDF up to 50 MB. ~200 tokens per page.

— —
Extracting layout-aware Markdown… ~5-10 sec/page
Ko ngā kōwhiringa hōhonu
Whakamutunga
Kua iti te tohu. Get More Tokens
Want better results? Kāhua Premium (GPT-5, Claude, Gemini) deliver higher quality. View Plans

❤️ E hiahia ana ki te Free.ai? Whakapāpāho ki ōna hoa!

Kaituhi ki kia whiwhi ai i tētahi pātahitanga tohutoro, ā, ka whiwhi i ngā tohu 25,000 ia hoa.

E hiahiatia ana ētahi atu? Ka whakaingoatia te wāteatanga mo ngā tohu 5K / rā + tāpiri 10K
Ka whakaingoatia te pūkete

E whakapāpā ana i tō tātau tono...

Convert any PDF into clean GitHub-flavored Markdown with headings, tables, lists, and code blocks preserved. Powered by IBM Granite-Docling. Free, unlimited, no signup.

He pēhea te whakamahi PDF to Markdown

1
Kei roto i tō tou tāuru

Type i te kupu, tuku i tētahi faila, whakaahua rānei i te mea e hiahiatia ana e koe. Kāore he tatau e hiahiatia ana.

2
Ka tirohia te whakatūnga

Ka tukatuka tātau AI i tō tātau tono i roto i ngā wā kotahi mā te whakamahi i ngā tauira pūtake tūwhera pai rawa.

3
Whakahua & tiritiri

Whakataki, tārua, tiritiri rānei i tōna hua. Whakatika noa iho mō te whakamahinga whaiaro, hokohoko rānei.

Ka whakamahia tēnei utauta mā te API

Ka whakamātautau tēnei utauta mai i tōtou waehere. Ko te wāhi mutunga o te REST e ōrite ana ki te OpenAI, te mana tohu-tokona, kāore he SDK tāpiri e hiahiatia ana. Ko ngā utu tohu e ōrite ana ki te whakawhitinga whatunga.

Ka taea te whakataki i te papatono Kitenga te kī API 
curl -X POST https://api.free.ai/v1/chat/ \
  -H "Authorization: Bearer sk-free-..." \
  -H "Content-Type: application/json" \
  -d '{"model": "qwen7b", "messages": [{"role": "user", "content": "Use the PDF to Markdown tool on: ..."}]}'

E pā ana ki ngā utauta AI wātea

PDF Extractor
PDF to Text
Academic Paper Extractor
Research Paper to Text
arXiv PDF Extractor

PDF to Markdown — FAQ

Drop in any PDF and the AI converts it into clean GitHub-flavored Markdown — headings stay headings, tables stay tables, lists stay lists, code blocks stay code blocks. Goes way beyond plain text extraction; the document's structural hierarchy is preserved so you can drop the output straight into a docs site, an LLM RAG pipeline, or a search index.

IBM Granite-Docling-258M (Apache 2.0). Tiny vision-to-sequence model fine-tuned for layout-aware document conversion — beats pdftotext + much faster + smarter than running a generic vision-language model on each page.

pdftotext is a flat dump — paragraphs and tables collapse into a wall of words. Adobe Export to Word preserves layout but produces .docx + costs ~$15/mo. Docling preserves the SEMANTIC structure (heading levels, lists as lists, tables as Markdown tables) and outputs a format LLMs and dev tools can both consume natively.

LlamaParse and unstructured both have free tiers but cap pages/month and require an API key. Docling-258M runs locally on our GPU + is fully self-hosted Apache 2.0, no per-page metering, no key signup. Quality is competitive with LlamaParse on standard documents.

Yes — tables come back as proper Markdown pipe-tables. Complex multi-column / nested tables are flattened more aggressively (a fundamental Markdown limitation, not the model's fault). For perfect table fidelity, we also support `format=html` via the API which preserves rowspan/colspan.

Granite-Docling does the OCR step itself — works on born-digital AND scanned PDFs alike. Scanned at lower DPI (<150) loses some text accuracy; rescan at 200+ DPI for best results.

Most LaTeX-rendered equations come through as inline `$...$` Markdown math. For research papers with heavy math, we also offer the academic-paper-extract tool (Nougat) which is specifically tuned for equations and citations.

About 5-10 seconds per page on our H200. A 30-page report is ~3-5 minutes. Tiny model means batches of small PDFs are essentially free in the daily pool.

200 tokens per page, with a 500-token floor. A 5-page contract = 1,000 tokens. A 30-page report = 6,000 tokens. The 5K daily free pool covers most typical use.

PDF — born-digital + scanned both supported. Max 50 MB upload. Other document formats (DOCX, EPUB, HTML, etc.) are on the roadmap; for now upload-and-convert with the pdf-conversion tool first.

Processed immediately, the Markdown output is kept (24h anonymous / 7d paid share-link expiry), the source PDF is deleted right after extraction. Never used for training. /privacy/ for the full policy.

Yes — POST a multipart `file` to /v1/document/pdf-to-markdown/. Returns {markdown_url, pages, preview, tokens, share_url}. Bearer auth (sk-free-…) gives 10K free tokens/month. /api/ has the curl example.

Ka whakaingoatia te wāteatanga mō ngā tohu 10,000

Ka waihanga tētahi pūkete wātea

Kāore he kāri ā-pūtea e hiahiatia ana

He pēhea te whakawātea i tēnei utauta?

E hiahia ana ki te Free.ai? Whakapāpāho ki ōna hoa!