#6565th of 168 in ML frameworks and librariesFor finance and business, trading, stocks and crypto, finance and accounting
qlib
Python platform for ML-based quantitative investment research and backtesting
- Stars
- 50.2k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
40,614 on 20 Apr 202650,216 on 9 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Qlib is a Python library covering the quant research pipeline: data handling, model training, backtesting, portfolio optimization and order execution. It ships a zoo of paper-implemented forecasting models (LightGBM, TCN, Transformer, TabNet, HIST, TRA and others), plus meta-learning and reinforcement learning frameworks. Workflows run from YAML configs via qrun or are built in code.
Who it is for: Quant researchers building ML-driven trading strategies in Python
Strengths
- Covers data, modeling, backtesting, portfolio optimization and execution in one codebase
- Includes many published quant models, with example configs run through qrun
- Components are loosely coupled and can be used stand-alone
- Supports supervised, market-adaptation (DDG-DA) and reinforcement learning workflows
Weaknesses
- Official dataset is disabled for now; README points to a community data source
- Bundled data comes from Yahoo Finance and the README says it may be imperfect
- Unreleased main-branch change restricts recorder loads and needs trusted=True opt-in
- Examples target China A-share data; other markets need your own data
What it needs
- Docker
- Needs Python 3.8-3.12, Yahoo Finance or community market data
- Models: LightGBM, TCN, Transformer, TabNet, HIST
Quick answers
Is qlib open source?
Yes. Its code is on GitHub under the MIT license.
Can you run qlib with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Also in ML frameworks and libraries
See all 168| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |