Home / LLM agents & research
pipiku915/FinMem-LLM-StockTrading
LLM stock trading agent with layered memory
What it is
A research framework for an LLM-based stock trading agent called FinMem. Its central idea is a layered memory that keeps market information at different time depths, combined with a configurable character profile that shapes how the agent makes decisions. Researchers and agent developers who want to reproduce or extend memory-based LLM trading experiments can build on it. Results depend heavily on the language model and data you plug in, so treat it as an experiment framework rather than a ready strategy.
At a glance
Research onlyOur rating, based on popularity, maintenance and how ready it is for real use.
| Best for | Professional quants |
|---|---|
| Used for | Strategy research, AI agents |
| Markets | US equities |
| Stack | Python |
| Learning curve | Moderate learning curve |
| Practical value | Medium practical value |
| Cost | Free and open source |
| Hardware | No GPU needed |
| Maintenance | No commits in over six months |
GitHub stars, last 30 days
Daily snapshots since 2026-09-28 (up to 30 days): +7 over the period, now 967. Gaps mean no snapshot was taken that day.
FinMem: A Performance-Enhanced LLM Trading Agent with Layered Memory and Character Design
Similar tools
LLM trading research agent with multi-market backtesting
Open-source LLM harness for agentic trading
Multi-agent LLM research system for China A-shares
Reinforcement learning agent for stock trading research
Python design patterns for building an AI stock trading bot
LLM multi-agent investment research framework specialized for A-shares