Chapter 4 · GitHub facts 2026-10-22
FinRL-Trading
What it does
The README titles the project FinRL-X and describes it as modular infrastructure for quantitative trading that unifies data processing, strategy composition, backtesting, and broker execution through a weight-centric interface. It links a paper at arxiv.org/abs/2603.21330. requirements.txt includes torch, lightgbm, xgboost, yfinance, alpaca-py, openai, and streamlit. This list files it as that FinRL-X stack.
Repository facts
| Repository | AI4Finance-Foundation/FinRL-Trading |
| GitHub description | FinRL-X: An AI-Native Modular Infrastructure for Quantitative Trading |
| Stars | 3,845 |
| Forks | 1,123 |
| Created | 2020-07-26 |
| Default branch | master |
| Last commit | 2026-09-18 4409abe |
| Last push | 2026-09-18 |
| License | Apache-2.0 |
| Homepage | https://ai4finance.org |
| Primary language | Python |
| Languages | Python, Shell, Dockerfile |
| Topics | deep-reinforcement-learning, stock-trading, a2c-algorithm, ppo, ddpg, ensemble-strategy, stock-trading-strategy, automated-stock-trading, sharpe-ratio, finrl, portfolio, portfolio-allocation, stock-selection, finrl-x |
Source: GitHub API, 2026-10-22.
Technical stack
Primary language reported by GitHub: Python. Languages reported by GitHub: Python, Shell, Dockerfile.
Dependency files (package names only, as read on the fetch date):
requirements.txt:numpy,pandas,scikit-learn,lightgbm,xgboost,yfinance,alpaca-py,openai,streamlit,torch,finnhub
Activity
Last commit on the default branch master: 2026-09-18 (4409abe). Subject: Merge pull request #95 from emt-Q/fix/ml-strategy-apply-risk-limits.
Last push to the repository: 2026-09-18. Created: 2020-07-26.
Who it is for
For people moving from the original FinRL tutorials to the modular stack. The FinRL README points at this repository for the newer architecture.
In this list
Filed under: Chapter 4.
Category note: Trading and Portfolio.
Related project pages: FinRL, Qlib.
Comparisons: Best open-source trading agents.