whatisgithub

What is arxiv-sanity-lite?

karpathy/arxiv-sanity-lite — explained in plain English

Analysis updated 2026-07-17 · repo last pushed 2023-06-19

★ 1,637PythonAudience · researcherComplexity · 2/5DormantSetup · moderate

In one sentence

A lightweight web app that learns which arxiv papers you like from tags you give it, then recommends similar new papers so you don't have to browse arxiv manually.

Mindmap

mindmap
  root((repo))
    What it does
      Tags papers you like
      Recommends new papers
      Daily email digests
    Tech stack
      Python
      SVM text model
      Local database
    Use cases
      Track research field
      Filter arxiv noise
      Email paper digest
    Audience
      Researchers
      Grad students

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Tag interesting papers and get automatic recommendations for similar new arxiv submissions.

USE CASE 2

Search, filter, and sort arxiv papers through a clean web interface instead of arxiv's own site.

USE CASE 3

Receive a daily email digest of papers matching your research interests without logging in.

USE CASE 4

Self-host a lightweight paper-tracking tool on a cheap $5/month server for a research group.

What is it built with?

PythonSVMSQLite

How does it compare?

karpathy/arxiv-sanity-litewxyhgk/retain-pdfvasu-devs/justhireme
Stars1,6371,6541,703
LanguagePythonPythonPython
Last pushed2023-06-19——
MaintenanceDormant——
Setup difficultymoderateeasymoderate
Complexity2/53/53/5
Audienceresearcherresearchervibe coder

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 1h+

Requires setting up a background job to regularly pull new papers from arxiv and running your own small server.

Copy-paste prompts

Prompt 1
Help me deploy arxiv-sanity-lite on a small $5/month server to track papers in my research area.
Prompt 2
Explain how arxiv-sanity-lite's SVM-based recommendation system decides which new papers match my tagged interests.
Prompt 3
Show me how to set up the daily email digest feature in arxiv-sanity-lite.
Prompt 4
Walk me through customizing arxiv-sanity-lite's search to filter papers by a specific arxiv category.
Prompt 5
How would I scale arxiv-sanity-lite's local database and scan-based search if I wanted to index 100,000 papers instead of 30,000?

Frequently asked questions

What is arxiv-sanity-lite?

A lightweight web app that learns which arxiv papers you like from tags you give it, then recommends similar new papers so you don't have to browse arxiv manually.

What language is arxiv-sanity-lite written in?

Mainly Python. The stack also includes Python, SVM, SQLite.

Is arxiv-sanity-lite actively maintained?

Dormant — no commits in 2+ years (last push 2023-06-19).

How hard is arxiv-sanity-lite to set up?

Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.

Who is arxiv-sanity-lite for?

Mainly researcher.

Open on GitHub → Ask about another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.

What are similar repos?