Home › Language Models › FinGPT

FinGPT - AI Language Models Tool

Overview

FinGPT is an open-source project from the AI4Finance Foundation that provides ready-to-use financial large language models (LLMs), training notebooks, fine-tuning pipelines (LoRA / QLoRA / 8-bit workflows), benchmark datasets, and demo applications such as a FinGPT-Forecaster robo-advisor. The repository bundles multi-task and single-task finetuned models (examples include Llama-2, Falcon, Bloom, MPT, Qwen, ChatGLM2 variants) and published instruction-tuning datasets for sentiment analysis, relation extraction, headline classification, NER and financial Q&A. ([raw.githubusercontent.com](https://raw.githubusercontent.com/AI4Finance-Foundation/FinGPT/master/README.md)) Designed for researchers and practitioners who need cost-effective adaptation of base LLMs to financial tasks, FinGPT emphasizes lightweight fine-tuning (LoRA/QLoRA) and retrieval-augmented approaches (RAG) to incorporate time‑sensitive market information. The project publishes models and dataset artifacts on Hugging Face and provides example notebooks that reproduce benchmarks and deploy demos (including a Hugging Face Space for the FinGPT-Forecaster). FinGPT is released under an MIT license and distributed via the GitHub repository and a small PyPI package. ([raw.githubusercontent.com](https://raw.githubusercontent.com/AI4Finance-Foundation/FinGPT/master/README.md))

GitHub Statistics

  • Stars: 18,660
  • Forks: 2,617
  • Contributors: 29
  • License: MIT
  • Primary Language: Jupyter Notebook
  • Last Updated: 2026-02-11T12:29:30Z

Repository activity and community: FinGPT is actively maintained and widely used — the GitHub repository shows ~18.7k stars, ~2.6k forks and 30 contributors, with ~77 open issues (indicating ongoing community engagement and troubleshooting). Recent commits show continued maintenance into 2026 (commits on Feb 11, 2026 and other updates through late 2025 and early 2026). The README and notebooks are the primary user-facing artifacts (Jupyter notebook-heavy codebase). ([github.com](https://github.com/AI4Finance-Foundation/FinGPT)) Health indicators: strong star/fork counts and multiple contributors signal broad interest; the relatively high issue count and third‑party reports highlight installer/compatibility friction (dependencies like bitsandbytes/accelerate and Hugging Face demo funding have been discussed in issues), so expect some environment/setup troubleshooting before running large-model flows. ([thedispatch.ai](https://thedispatch.ai/reports/1904/?utm_source=openai))

Installation

Install via pip:

pip install FinGPT
git clone https://github.com/AI4Finance-Foundation/FinGPT.git
cd FinGPT
python -m venv venv && source venv/bin/activate  # or venv\Scripts\activate on Windows
pip install -r requirements.txt  # repository includes requirements.txt (may be minimal); some notebooks expect additional ML libraries
pip install -e .  # optional: install package in editable mode
jupyter lab  # open the provided Jupyter notebooks (notebooks demonstrate LoRA/QLoRA training and inference)

Key Features

  • LoRA and QLoRA example notebooks for low-cost finetuning of base LLMs.
  • Published instruction-tuning datasets for sentiment, NER, relation extraction, and QA.
  • FinGPT-Forecaster demo (Hugging Face Space) for short-term stock movement analysis.
  • Retrieval-augmented (RAG) pipelines specialized for financial sentiment and context.
  • Multi-model releases: LoRA adapters for Llama-2, Falcon, Bloom, MPT, Qwen, ChatGLM2.

Community

Active open-source community: ~18.7k GitHub stars, ~2.6k forks and ~30 contributors. Discussion and issues are active (≈77 open issues); recent commits through Feb 11, 2026 show ongoing maintenance. Community feedback highlights strong interest but also reports setup and model-integration challenges (dependencies, demo funding, dataset generation errors) that are tracked in issues and third-party analyses. Recommended: join the project's Discord and GitHub discussions for setup help and updates. ([github.com](https://github.com/AI4Finance-Foundation/FinGPT))

Last Refreshed: 2026-02-24

Key Information

  • Category: Language Models
  • Type: AI Language Models Tool
Visit Official Website

Related Tools

Qwen2.5-7B

Qwen2.5-7B is a large language model designed for text generation, featuring improvements in coding, mathematics, instruction following, long text generation, and multilingual support. It supports context lengths up to 128K tokens and is intended for sophisticated NLP tasks.
DeepSeek-V3

Open‑weight flagship LLM (base and chat) with strong reasoning and tool‑use; runs with SGLang/LMDeploy/vLLM and permissive model license.
Llama 3

Llama 3 is an open access large language model (LLM) released by Meta, available in various configurations (8B and 70B parameters) with capabilities for fine-tuning and integrations into platforms like Hugging Face, Google Cloud, and Amazon SageMaker.
UNfilteredAI-1B

A large-scale text generation model designed for creative and unconstrained content production without traditional filtering.
Shuttle-3

Shuttle-3 is a state-of-the-art language model designed for high-quality text generation, particularly suited for complex chat, multilingual communication, and reasoning tasks. It is fine-tuned from the Qwen-2.5-72b-Instruct model and designed to emulate high-quality prose similar to Claude 3 models.
WizardLM

WizardLM is a state-of-the-art large language model designed for complex chat, multilingual tasks, reasoning, and agent functionalities. It features an AI-powered pipeline (Auto Evol-Instruct) that optimizes instruction datasets for improved performance across various domains and leverages Arena Learning for an expanded learning pool of challenging instruction data.