arXiv PDF Extractor
商业用途OK
380+模型
无水印
不需要注册
型号 :
+ GPT-5, Claude, Gemini
Drop an arXiv preprint, journal paper, or thesis chapter — AI converts it into clean LaTeX-flavored text. Math equations stay as equations, multi-column layouts get unwound, citations preserved. Powered by Meta Nougat-base.
Reading equations + unwinding columns… ~10 sec/page
高级选项
结果成果成果成果成果成果成果成果成果成果成果
声调越来越低
Get More Tokens
❤️ Love this tool? Share it!
< a href=" "/ signup/" style=" "color:#16A34A" > 签名 以获得查询链接, 并获得每个朋友25,000个象征性的 。
还要吗?
每日5K象征性的免费签名 + 10K奖金
签署自由
处理您的请求...
Drop an arXiv preprint, get clean LaTeX-flavored text with every equation rendered inline. Multi-column layouts handled, references kept intact. Free, AI-powered.
如何使用 arXiv PDF Extractor
1
输入输入
键入文本、上传文件或描述您想要的东西。不需要账户 。
2
点击生成
我们的人工智能使用最佳的开放源码模型,在秒内处理你的请求。
3
下载共享( S)
下载、复制或分享您的结果。 免费个人和商业使用 。
通过 API 使用此工具
从您自己的代码中自动启用此工具。 OpenAI 兼容的 REST 端点、 Bearer-tok 异常点、 不需要额外的 SDK 。 Token 成本符合网络界面 。
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 arXiv PDF Extractor tool on: ..."}]}'arXiv PDF Extractor — FAQ
Drop in an arXiv preprint and the AI converts the entire paper into clean LaTeX-flavored text. Equations come back as proper LaTeX, multi-column layouts unwound, references intact. Built on Meta Nougat, trained specifically on millions of arXiv pages.
Nougat's training corpus was arXiv preprints — so it absolutely shines on the IEEE / ACM / NeurIPS / ICML / arXiv layout family. Other PDF extractors choke on multi-column math; this one was designed for it.
Download the PDF from arXiv (e.g. arxiv.org/pdf/2401.12345), upload it here, get back a single .txt file with the full paper as LaTeX-flavored text. No arXiv API key needed; we just need the PDF.
Yes — that's the headline feature. Inline math is `$...$`, displayed math `$$...$$`. Even raster-rendered equations in older papers come through correctly because the model treats each page as an image.
Auto-handled. Two-column IEEE-style is the most common arXiv layout and Nougat unwinds it into proper reading order without a config flag.
Yes — inline `[12]` / `[Smith2020]` markers stay where they belong, and the full reference list at the end is extracted intact for downstream BibTeX / Zotero use.
~8-15 sec/page. A 12-page conference paper takes ~2-3 min. NeurIPS-style 30+ page papers with appendices: 8-12 min. Submit and walk away.
300 tokens/page, floor 600. Most arXiv conference papers (8-15 pages) are 2,400-4,500 tokens. Daily free pool covers ~1-2 papers/day for signed-in users; paid plans get unlimited.
Feed it to ChatGPT / Claude for "explain this paper", build personal RAG over your saved papers, semantic-search your reading list, copy equations into your own LaTeX project, or read the paper as plain text on your phone.
Yes — Nougat OCRs internally. arXiv has been LaTeX-rendered for 25+ years so most preprints are clean digital. Older scanned papers work but math fidelity drops slightly; rescan at 300+ DPI for best results.
PDFs deleted right after extraction. LaTeX output is kept 24h (anonymous) / 7 days (paid share link). Never used for training. arXiv PDFs are public CC-BY anyway, but we don't store them either way.
Yes — POST multipart `file` to /v1/document/academic-pdf/. JSON response with `text_url`, `pages`, `preview`, `tokens`, `share_url`. Bearer auth (sk-free-…) gives 10K free tokens/month. /api/ for curl example.
你会如何评分这个工具?