---
id: qlib-paper
name: "Qlib: An AI-oriented Quantitative Investment Platform"
summary: The Microsoft Research preprint introducing Qlib, cited by the Qlib repository itself as the project's reference paper.
authors: [Xiao Yang, Weiqing Liu, Dong Zhou, Jiang Bian, Tie-Yan Liu, Microsoft Research]
origin: external
repo: https://github.com/microsoft/qlib
license: unknown
year: 2020
venue: arXiv preprint
arxiv: "2009.11189"
tags: [quant-platform, infrastructure, machine-learning, alpha-research, preprint]
markets: [equities]
added: 2026-08-17
prerequisites:
  trading: moderate
  ai: moderate
  programming: none
  setup: none
---

> Part of [Athenara](https://athenara.info/), the open registry for agentic trading.
> Index of everything: https://athenara.info/llms.txt · This entry: [HTML](https://athenara.info/papers/qlib-paper/) · [JSON](https://athenara.info/api/v1/papers/qlib-paper.json)

The preprint argues that adopting AI in quantitative investment demands an infrastructure upgrade,
and presents Qlib as that infrastructure — a platform meant to "realize the potential, empower the
research, and create the value of AI technologies in quantitative investment" across the full
pipeline of alpha seeking, risk modeling, portfolio optimization, and order execution.

It is a single-version arXiv submission from 22 September 2020, classified q-fin.GN with cs.LG and
q-fin.PM, and was never published to a conference or journal — there is no journal reference and
no DOI beyond arXiv's own, and the abstract page declares no reuse license for the text. It is
short and descriptive rather than an evaluation paper, and it documents the 2020 design: the
repository has since moved well past it, advertising later work the paper does not cover.

Its value is largely connective. The Qlib README names this arXiv entry as the project's own
reference paper, so a reader arriving at either artifact can reach the other, and the software it
describes — MIT-licensed, last pushed July 2026 — is the entry indexed here as an architecture.