FinGPT

Open-Source Financial
AI Platform

Build, evaluate, and deploy financial AI applications with open-source models, APIs, benchmarking, and enterprise services.

Train and adapt financial models at a fraction of the cost of traditional large-scale systems.

DownloadView DocsGitHub
20K+
GitHub Stars
500K+
Open-source Downloads
120+
Contributors
50+
Countries
30+
Papers
Used by researchers fromColumbiaStanfordMITNYUCMUOxfordNUSHKUST

What You Can Build with FinGPT

A financial AI platform for research, developer workflows, and real-world deployment.

Sentiment Analysis

NLP

Analyze financial news, filings, and market narratives with domain-tuned models and APIs.

Monitor real-time market sentiment across news sources and social media.

Forecasting

Prediction

Generate forecasting signals for financial research, market analysis, and model experimentation.

Predict stock price movements using AI-powered analysis.

Try Demo

Retrieval for Finance

RAG

Ground responses in filings, reports, and structured financial data with retrieval-augmented workflows.

Query SEC filings and earnings reports with natural language.

Multi-Agent Analysis

Agents

Coordinate multiple AI agents across sentiment, market data, and fundamentals for deeper analysis.

Run coordinated analysis across multiple data sources simultaneously.

Report Intelligence

Documents

Extract metrics, risks, and insights from earnings reports, annual filings, and financial documents.

Automatically extract key metrics from quarterly earnings reports.

Benchmarking

Academic

Evaluate finance-specific LLM performance across domain datasets, tasks, and practical workflows.

Compare model performance on standardized financial AI tasks.

API Access

Developer

Integrate FinGPT capabilities into applications and research pipelines through developer-friendly APIs.

Add sentiment analysis to your trading pipeline with a single API call.

Enterprise Services

Enterprise

Enable custom deployment, model customization, support, and security controls for organizational use.

Deploy FinGPT on-premise with dedicated support and custom models.

Open Source at the Core, Scalable for Production

FinGPT provides open-source financial AI models, managed APIs, benchmarking tools, and enterprise deployment support for teams building real-world applications.

FinGPT Demo

Enter any ticker symbol, and get AI-powered stock movement predictions with news analysis, sentiment scoring, and financial summaries.

fingpt.io/forecaster

Companies from Dow-30 are recommended

AAPL

Prediction start date (yyyy-mm-dd)

2024-01-15

Past weeks of data to analyze (1–4)

3
Use Latest Basic Financials
[Company Introduction]

Apple Inc is a leading entity in the Technology sector. Market cap of $2.83T USD with 15,634M shares outstanding. Trading under AAPL on NASDAQ.

[Price Movement]

From 2024-01-08 to 2024-01-15, AAPL's stock price decreased from $185.84 to $183.63.

[Recent News]

Apple Vision Pro pre-orders exceed expectations

Strong demand signals positive revenue outlook for Q2...

iPhone 15 sales beat estimates in holiday quarter

Analysts raise price targets following robust Q1 results...

[Prediction]
Next week outlook:UP 2.3%

* This is a demo showing model capabilities. Not financial advice. Always consult a professional before trading.

Ecosystem

A connected open-source ecosystem spanning financial language models, reinforcement learning, trading systems, and AI agents.

FinGPT

Open-source financial AI platform for language models, benchmarking, APIs, and research workflows.

FinRL

Deep reinforcement learning framework for quantitative finance and portfolio research.

FinRL-Trading

AI-native trading infrastructure for strategy development, backtesting, and live execution.

FinRobot

Agent-based framework for financial analysis, research automation, and task orchestration.

Explore the Ecosystem

Benchmarks

FinGPT benchmarks demonstrate domain performance, cost efficiency, and practical readiness for financial AI applications.

Zero-shot Evaluation on Financial PhraseBank

ModelAccuracyF1
Instruct-FinGPT0.760.74
ChatGPT 4.00.640.51
Llama-7B0.600.40
ChatGLM2-6B0.470.40
BloombergGPT—0.51

Training Cost Comparison Between LLMs

ModelDevicesTimeCost
Instruct-FinGPT8 × A1002 hours$65.6
ChatGLM264 × A1002.5 days$14,976
BloombergGPT512 × A10053 days$2.67M
Llama32048 × A10021 days$4.23M

* Based on $4.1 per GPU per hour (A100)

FinGPT training cost:$65.6vs$2.67M(BloombergGPT)vs$4.23M(Llama3)
Higher Performance
Domain-tuned models outperform general-purpose LLMs on financial tasks
Lower Training Cost
Fine-tune for ~$65 vs millions for training from scratch
Better Adaptation
Instruction tuning enables rapid adaptation to new financial datasets
Production Ready
Strong fit for both research exploration and production deployment

Download

Open-source models, source code, and release packages — free under the MIT license.

GitHub Repository

Full source code, training scripts, and documentation.

HuggingFace Models

Pre-trained and fine-tuned model weights ready to download.

Releases

Versioned packages, changelogs, and release notes.

View All Downloads

Start Building with FinGPT

Explore open-source models, integrate APIs, or deploy enterprise solutions for financial AI.

Download FinGPTGitHub