# FinGPT

Open-source financial large language model project for finance sentiment, analysis, and domain adaptation.

## Agent Decision Summary
- Risk level: low
- Source confidence: high
- Recommended workflows: Coding agent workflow
- Permission surface: low explicit permission surface in metadata
- Agent JSON: https://www.openagent.bot/models/fingpt.agent.json

## Summary
FinGPT is an open-source financial LLM project from AI4Finance Foundation. It focuses on finance-specific language modeling, sentiment analysis, and domain workflows where general LLMs often need adaptation.


## Guide
FinGPT is an open-source financial LLM project from AI4Finance Foundation.

### What it is
It focuses on financial language modeling, sentiment analysis, and finance-specific workflows.

### Why it matters
Finance is a high-stakes domain where open methods, evaluation, and data choices matter.

### How it works
Use FinGPT as a research baseline, then validate it on your own finance tasks with strict evaluation and compliance review.


### FAQ
- Is FinGPT open source?
  - Yes. The GitHub repository is listed under the MIT license.
- Can FinGPT provide financial advice?
  - FinGPT is a research and development project. Any financial use should go through expert review, compliance checks, and domain-specific evaluation.
## What It Does
It focuses on financial language modeling, sentiment analysis, and finance-specific workflows.

## How To Evaluate
Use FinGPT as a research baseline, then validate it on your own finance tasks with strict evaluation and compliance review.

## Why It Matters
Finance is one of the clearest domains where generic LLM behavior needs specialized data, evaluation, and governance. FinGPT gives teams an open project to study financial LLM workflows.


## Best For
- Researchers studying finance-focused LLMs
- Teams prototyping sentiment or financial text analysis
- Builders comparing domain-specific open-source model projects

## Not For
- Teams looking for financial advice or trading guarantees
- Users who need a general-purpose agent framework

## What It Actually Does
- Finance-specific scope: FinGPT focuses on financial text, sentiment, and domain workflows.
  - Why it matters: Domain models can be easier to evaluate than generic models when the task has specialized vocabulary and risk.
- Open research project: The project is maintained openly by AI4Finance Foundation.
  - Why it matters: Researchers can inspect the approach and adapt it for finance-focused experiments.
- Practical task orientation: FinGPT is relevant for sentiment, robo-advisor, and market text analysis experiments.
  - Why it matters: Finance teams need task-specific tests instead of broad benchmark claims.

## Typical Use Cases
- Financial sentiment analysis: Evaluate financial news, filings, and social text with domain-aware models.
- Domain adaptation research: Study how financial corpora affect model behavior.
- Finance agent backends: Prototype a finance assistant with an explicitly finance-oriented model layer.

## How It Compares
- When to choose FinGPT: Compare it with nearby models by looking at hosting model, integration surface, license, and whether the official docs show the workflow you need.

## Fit Matrix
- Coding agent workflow: strong. FinGPT has multiple signals for coding agent workflow, including matching tags, capabilities, category, or positioning. Required check: Run a small repository change and inspect the diff, tests, and rollback path.
- Browser automation: partial. FinGPT has at least one signal for browser automation, but should be checked against a real task before adoption. Required check: Run one non-sensitive website task and inspect clicks, waits, retries, and changed URLs.
- Local or private AI stack: partial. FinGPT has at least one signal for local or private ai stack, but should be checked against a real task before adoption. Required check: Verify hardware requirements, data path, storage, and whether all calls stay in your environment.
- Reusable skill workflow: partial. FinGPT has at least one signal for reusable skill workflow, but should be checked against a real task before adoption. Required check: Run one skill end to end and check whether it produces evidence or structured output.
- Connector or protocol layer: weak. FinGPT is not primarily positioned for connector or protocol layer in the current metadata. Required check: Connect one low-risk service, then inspect schemas, auth scope, errors, and logs.
- Evaluation and observability: weak. FinGPT is not primarily positioned for evaluation and observability in the current metadata. Required check: Add one repeatable test case and confirm results can run again in review or CI.

## Evidence
- verified: FinGPT is listed as open source. Source: License metadata: MIT
- verified: FinGPT has a recorded GitHub repository: AI4Finance-Foundation/FinGPT. Source: Resource facts and GitHub source link.
- inferred: FinGPT supports these recorded deployment modes: cloud. Source: OpenAgent decision signal metadata.
- inferred: FinGPT is tagged with local inference capabilities. Source: OpenAgent capability taxonomy.

## Missing Checks
- Dedicated docs link is missing.
- Repository freshness has not been recorded.

## Next Actions
- Inspect repository: https://github.com/AI4Finance-Foundation/FinGPT
- Open Homepage: https://fingpt.io

## Facts
- Category: models
- Resource type: model
- Open source: yes
- License: MIT
- Last verified: 2026-06-09
- GitHub repo: AI4Finance-Foundation/FinGPT
- GitHub stars: 20446

## Capabilities
- local-inference

## Structured Use Case Tags
- developer-workflow

## Getting Started
- Open the GitHub repository: https://github.com/AI4Finance-Foundation/FinGPT
- Visit the project website: https://fingpt.io

## Links
- GitHub: https://github.com/AI4Finance-Foundation/FinGPT
- Homepage: https://fingpt.io

## Structured Outputs
- JSON: https://www.openagent.bot/models/fingpt.json
- Markdown: https://www.openagent.bot/models/fingpt.md
- Agent JSON: https://www.openagent.bot/models/fingpt.agent.json
- Canonical: https://www.openagent.bot/models/fingpt