Insightconsultance.ai

OpenBB Review in 2026

8 min read

Muhammad AliMuhammad AliFounder, consultance.ai
OpenBB Review in 2026

OpenBB is a free, open source financial data platform (the Open Data Platform, installed with pip install openbb) that now ships under the Apache 2.0 license. OpenBB, Inc. announced its wind down on August 25, 2026, and the code is now stewarded by the non profit OpenBQ, with Workspace headed to FINOS. Use it as self run research and data infrastructure, not as a hosted product with vendor support.

What this covers

  • OpenBB's GitHub repo had 75,189 stars and 7,859 forks on November 7, 2026.
  • Version 5.0.0 moved the license from AGPL 3.0 to Apache 2.0 (PR #7677, merged September 24, 2026).
  • OpenBB, Inc. announced its wind down on August 25, 2026; the non profit OpenBQ now stewards the code.
  • The Workspace codebase is entering FINOS Labs, and pro.openbb.co now redirects to a BrightQuery waitlist.
  • ODP exposes data through Python, a local REST API and an MCP server for AI agents.
  • The software is free; data licenses, model API calls and engineering time are the real costs.

OpenBB is an open source financial data toolkit for analysts, quants and AI agents. Its main piece is the Open Data Platform (ODP), which you install with pip install openbb. It ships under the Apache 2.0 license as of version 5.0.0 in September 2026. The most important fact for anyone evaluating it in late 2026 is that OpenBB, Inc. announced its wind down on August 25, 2026, and the code is now stewarded by a non profit called OpenBQ.

That changes the review. The software is still free and still maintained, and the license is now friendlier to commercial use than it was. But there is no longer a vendor behind the hosted product, so you should treat OpenBB as code you run and own, not a service you subscribe to. This review covers what the project is, what it does, what the license allows, what it costs to run, and where it fits for a fund, a family office or a deal team.

What OpenBB is in 2026

The project began on December 20, 2020 as Gamestonk Terminal, written by founder Didier Lopes over a holiday in London, according to his own account on the OpenBB blog. That is the "OpenBB Terminal" many people still search for. Over the following years it grew into an SDK (now the Open Data Platform), the OpenBB Bot, OpenBB Workspace, OpenBB Copilot and an Excel add in.

The GitHub repository describes itself today as an "Open Data Platform for analysts, quants and AI agents." On November 7, 2026 it showed 75,189 stars and 7,859 forks, with commits pushed that same day. The README frames ODP as a "connect once, consume everywhere" layer: you wire up data sources once and expose them to Python, to OpenBB Workspace and Excel, to MCP servers for AI agents, and to REST APIs for other applications.

One practical detail: the familiar URL github.com/OpenBB-finance/OpenBB now redirects to github.com/openbq-org/OpenBB. The OpenBB blog confirms that the openbq-org GitHub organization belongs to OpenBQ, so the redirect is the expected result of the handover, not a fork you should be wary of.

Hands turning the pages of a tabbed diligence document under a desk lamp

AI for finance teams

Screen more deals without adding analysts

First system live in about six weeks. You own the code.

Book a discovery call

What happened to the company

In a post titled "OpenBB belongs to everyone," dated August 25, 2026, Lopes wrote that the company could not find the product market fit needed to build a sustainable business, and committed to releasing the full product suite under a permissive license. He named OpenBB Workspace, Open Data Platform, OpenBB Copilot and the Excel add in as part of that release, with support from investor OSS Capital.

An edit to that post dated October 1, 2026 adds three facts. OpenBQ is a non profit organization created to steward OpenBB technology, led by Jose Plehn from BrightQuery, who has hired a third of the OpenBB team. The Workspace codebase will be stewarded by FINOS, the fintech open source foundation under the Linux Foundation. And BrightQuery is building a hosted "BQ Workspace" on its own datasets, with a waitlist open.

The FINOS side is real and moving. The Labs contribution proposal for "FINOS Workspace," filed by BrightQuery as the FINOS member organization on September 14, 2026, is marked approved. It describes the Workspace source as Apache 2.0 licensed and still private at the time of submission. The hosted app at pro.openbb.co now redirects to workspace.brightquery.com.

What it actually does

The core is a Python library. The README example pulls historical prices in three lines: import obb, call a route, convert the result to a DataFrame. Each data source is a separate provider package, and each analytical area is a separate router package, so you install what you need.

The 5.0.0 metapackage on PyPI pulls in these providers: BLS, Cboe, CFTC, Deribit, ECB, Fama French, Federal Reserve, FINRA, FRED, US government data, IMF, JODI, Nasdaq, OECD, SEC, TMX and the US EIA. Routers for equity, ETF, fixed income, economy, crypto, currency, commodity, derivatives, index, quantitative, technical and econometrics analysis come through an optional routers extra. The V5 release notes say sixteen provider packages that existed on the development branch are absent from V5, so if a tutorial references a provider you do not see in that list, check before you plan around it.

Beyond the library, the package includes three surfaces worth knowing:

  • A REST API. Running openbb-api starts a FastAPI server at 127.0.0.1:6900, which is also how you connect ODP to OpenBB Workspace as a backend.
  • An MCP server. openbb-mcp-server exposes the installed routes to LLM agents through the Model Context Protocol. With tool discovery on, it hides the full catalog behind a few discovery tools so an agent searches for what it needs instead of loading every tool up front.
  • A CLI. openbb-cli is the descendant of the original terminal. V5 shipped CLI 2.0 with pluggable backends, spec files and code generation.

The license and what it allows commercially

This is the change that matters most for firms. Pull request #7677, merged on September 24, 2026, moved the repository from AGPL 3.0 to Apache 2.0, in the root LICENSE file, every package manifest and the desktop app. PyPI lists openbb 5.0.0 as Apache 2.0.

Under AGPL, a firm that modified the code and offered it to users over a network could be required to make the modified source available to those users. Apache 2.0 is a permissive license: you can use, modify and distribute the code, including inside proprietary internal systems, as long as you keep the license and notices and respect its patent and trademark terms. Older guides and some pages on openbb.co still describe ODP as AGPL, so check the version you actually install.

Two limits remain. The license covers the code, not the data: every provider has its own terms, and pulling a dataset through OpenBB does not change what you are allowed to do with it. And the README disclaimer states plainly that the data in the platform "is not necessarily accurate" and that OpenBB accepts no liability for reliance on it.

A shelf of neatly stacked bound reports in raking side light

Build library

See what we have already shipped

Working systems, the stack behind them, and what they cost to run.

Open the library

What it costs to run

Cost lineWhat the sources say
Software license$0. Apache 2.0, installed from PyPI.
InfrastructureRuns locally or on your own server. The PyPI package requires Python 3.10 or later. The API server runs on your own host.
DataBundled providers are mostly public sources. Some, such as FRED, require a free API key. Licensed market data comes from your own vendor contracts.
AI model callsNot bundled. The MCP server lets the agent and model of your choice call OpenBB, so model cost is whatever your provider charges.
Hosted WorkspaceThe openbb.co pricing page still lists Community (free), Lite ($2,400 per year team license), Pro (custom) and a Snowflake edition ($500 per year per seat). The hosted app now redirects to a BrightQuery waitlist, so treat those prices as historical until confirmed.
Engineering timeThe real cost. You install, run, upgrade and secure it yourself.

Find My Moat's October 2026 review reaches the same conclusion: the open source platform is free, but with the company wound down there is no vendor support or hosted product to rely on, and production datasets require your own providers.

Honest limits

OpenBB is a research and data integration tool. It is very good at getting public and licensed data into a DataFrame, an API or an agent quickly. It is not an order management system, an execution platform or a book of record, and nothing in the documentation claims it is.

The quantitative and technical routers support analysis and backtest style work on historical data. Moving from a backtest to anything live adds problems OpenBB does not solve for you: data entitlements, latency, reconciliation against your own positions, audit trails and model risk review. The README's own trading risk disclaimer is a fair summary of the posture you should take.

Governance is also in transition. The code has a new steward, Workspace is entering FINOS Labs, which is the entry stage of the FINOS project lifecycle, and the hosted product is being rebuilt by BrightQuery. None of that is a reason to avoid the project, but it is a reason to pin versions and read the changelog before each upgrade.

Who should use it, and who should not

Use it if you have Python skills on the team, want a free and permissively licensed way to pull macro, regulatory, filings and market data into one interface, and are comfortable running your own software. It is a strong starting point for a quant researcher, a data engineer building an internal research API, or a team that wants to give an AI agent governed access to financial data through MCP.

Skip it if you need a hosted product with a support contract, licensed real time market data included in the price, or a tool non technical analysts can adopt without engineering help. Right now those needs are better served by a commercial vendor or by waiting to see what BrightQuery and FINOS ship.

A quiet corner of a modern office at dusk with city light through the window

Implementation services

Put AI into the workflow you actually run

Diligence, close, reconciliation, reporting — in your environment.

See how it works

Where it fits for a fund, family office or deal team

For a private equity firm, OpenBB is most useful at the edges of the deal process: macro and sector context, public comparables data, SEC filings and regulatory data pulled into a model or memo. The core diligence work happens inside your own data room documents, CRM and portfolio data, which no public data toolkit reaches by itself.

For a family office, the appeal is cost and ownership. A small team can stand up a free research API over public data and point an internal agent at it. The gap is the private side: custodian feeds, capital account statements and manager reports still need their own ingestion and reconciliation.

The project's founder argued the same point in his farewell post: the future of financial software is firms building environments that reflect their own data, systems, processes and compliance requirements. OpenBB gives you a good open data layer for that environment. It does not give you the rest of it.

When an in environment build is the better answer

If your bottleneck is getting public data into a notebook, install OpenBB and stop there. If your bottleneck is a specific workflow, such as screening deals, drafting the first pass of financial due diligence, or reconciling statements across custodians, the data layer is the smallest part of the job. The work is connecting your private documents and systems, encoding your team's method, and putting an audit trail around every number.

That is what consultance.ai builds: custom finance AI that runs in your environment, with the first system live in about six weeks and the code owned by you. OpenBB can sit inside that build as one of the data sources. If you want to see what a workflow like that looks like, the consultance.ai library has working examples, or you can book a call to talk through yours.

Sources

Bring one workflow to a 30-minute audit.

Book an audit