# AI Research Agents & Copilots

Agent frameworks you run on your own model keys and research copilots you subscribe to — what a run costs, what comes out, and whether anything trades.

*https://stockmarketstack.com/categories/ai-agents*

Everything in this listing does the reading itself. You hand it a ticker or a question, a model
works through filings, prices, news or transcripts, and it comes back with a report. The job is the
same throughout; the delivery form decides most of what follows. A **framework** is open-source code
you run with your own model key and your own data keys: you pay per token, you can read every
prompt, and nobody supports it. A **copilot** is a hosted service you subscribe to: someone else
holds the corpus and the model, and what you can check is the citation.

- **Investor personas scoring US stocks, replayed as a fund** — [AI Hedge Fund](https://stockmarketstack.com/tools/ai-hedge-fund), MIT, on a paid Financial Datasets key.
- **Analysts, a bull, a bear and risk agents debating one ticker** — [TradingAgents](https://stockmarketstack.com/tools/tradingagents), Apache-2.0, keyless yfinance data by default.
- **An analyst-style report built from statements and a DCF** — [FinRobot](https://stockmarketstack.com/tools/finrobot), Apache-2.0, OpenAI and FMP keys.
- **A local research workspace with backtests and broker connectors** — [Vibe-Trading](https://stockmarketstack.com/tools/vibe-trading), MIT, about twenty model providers.
- **AI search over filings, broker research and expert transcripts** — [AlphaSense](https://stockmarketstack.com/tools/alphasense), enterprise contracts only.
- **Discontinued** — [Fintool](https://stockmarketstack.com/tools/fintool), bought by Microsoft on 18 April 2026.

## What decides the choice

**Which model, and what a run costs.** None of the frameworks ships a model. AI Hedge Fund takes
nine models across six providers, TradingAgents and Vibe-Trading a dozen or more providers each,
local Ollama included, and FinRobot's report generator is built on the OpenAI SDK. Nobody publishes
a cost per run, so reason from the structure. TradingAgents at default depth is four analyst legs,
two debate turns, three risk turns and three manager and trader calls, and its "Deep" preset makes
the two debate turns ten. AI Hedge Fund caches every prompt, so a re-run over unchanged filings
costs nothing. Vibe-Trading's own guidance is that small models fabricate answers instead of calling
tools: the cheap model is not the cheap run.

**What it reads.** AI Hedge Fund reads [Financial Datasets](https://stockmarketstack.com/tools/financial-datasets) and nothing
else, and its personas see fundamentals alone — no prices, no news. FinRobot reads Financial
Modeling Prep, whose free plan allows 250 calls a day. TradingAgents defaults to yfinance, and
Vibe-Trading to a fallback chain of 27 sources, most of them keyless public endpoints nobody
promised to keep. AlphaSense's contract buys a corpus no framework can reach: sell-side research and
Tegus expert transcripts among 500 million-plus documents.

**What comes out.** AI Hedge Fund: a conviction between −1 and +1 and a written thesis per persona,
blended by deterministic code into target weights and replayed in a backtest with no trading costs.
TradingAgents: a tree of Markdown reports ending on a rating from Buy to Sell, with an entry, a
stop, sizing and an optional price target. FinRobot: an HTML or PDF report whose cover rating and
target are the latest sell-side figures reprinted from FMP, with the model writing the thesis around
them. Vibe-Trading: a run card with metrics and a tool trace.

**Whether anything can trade.** Only Vibe-Trading, through eight brokers inside a mandate you commit
on symbols, size and exposure, with a kill switch. The project calls that path experimental, and
keeps its order tools off its MCP server. The other three frameworks execute nothing.

**Whether you can check it.** A copilot's answer is worth what its citation lets you verify.
AlphaSense cites inline, down to the source document; its "no hallucinations" line is marketing.
Fintool's located a document and a page, not proof of the arithmetic built on it. With a framework,
what you can audit is the prompt and the tool trace, and TradingAgents' own documentation says two
runs of the same ticker on the same date can disagree.

## How they fail

A data tool that breaks usually breaks where you can see it: an error, an empty table, a missing
field. A research agent that breaks writes a paragraph, as fluent as a good one, with nothing in it
to mark the figure the model invented or the filing it never opened — Vibe-Trading on a weak model
answers from training data without saying so. Read the trace or the citation before the conclusion.

Nothing here is endorsed. The ratings, convictions, price targets and "PM final call" these products
emit are generated output nobody stands behind, as the projects' own disclaimers say. This catalogue
describes how each one works and what it outputs, never which call to act on.

## What is not here, and where it went

[FinGPT](https://stockmarketstack.com/tools/fingpt) has moved: it is adapters, instruction datasets and training notebooks —
weights and data, not an application — and sits in [backtesting
frameworks](https://stockmarketstack.com/categories/backtesting-frameworks) under the research job, beside
[Kronos](https://stockmarketstack.com/tools/kronos), [FinRL](https://stockmarketstack.com/tools/finrl) and [Qlib](https://stockmarketstack.com/tools/qlib). A copilot inside a product
bought for another job keeps that job's category: [Fiscal.ai](https://stockmarketstack.com/tools/fiscal-ai),
[Daloopa](https://stockmarketstack.com/tools/daloopa) and [Quartr](https://stockmarketstack.com/tools/quartr) in [fundamentals
research](https://stockmarketstack.com/categories/fundamentals-research), [Bigdata.com](https://stockmarketstack.com/tools/bigdata-com) and
[AskNews](https://stockmarketstack.com/tools/asknews) in [news and sentiment APIs](https://stockmarketstack.com/categories/news-sentiment-apis). [AI stock
research and analysis tools](https://stockmarketstack.com/collections/ai-tools) sorts all of them by what the model does.

## Where this category connects

The data a framework runs on is bought in [market data APIs](https://stockmarketstack.com/categories/market-data-apis). Wiring a
model to data without a research loop around it is [financial MCP servers](https://stockmarketstack.com/categories/mcp-servers),
and [how to give an AI assistant market data](https://stockmarketstack.com/how-to/give-an-ai-assistant-market-data) is the
setup. [Financial MCP servers, and the licence behind them](https://stockmarketstack.com/guides/mcp-and-financial-data) covers
where models are weak with financial data: arithmetic over tables, point-in-time figures, and
fetched documents read as instructions.

## Cards

- [AI Hedge Fund](https://stockmarketstack.com/tools/ai-hedge-fund.md) — LLM investor personas score a ticker, deterministic code sizes the book, nothing trades.
- [AlphaSense](https://stockmarketstack.com/tools/alphasense.md) — AI search over 500M+ filings, broker research and Tegus expert transcripts.
- [FinRobot](https://stockmarketstack.com/tools/finrobot.md) — Open-source agents that turn a ticker and your own API keys into an analyst report.
- [Fintool](https://stockmarketstack.com/tools/fintool.md) — Discontinued — Microsoft bought the SEC-filings research copilot on 18 April 2026.
- [TradingAgents](https://stockmarketstack.com/tools/tradingagents.md) — Analyst, bull, bear, risk and portfolio-manager agents argue over one ticker.
- [Vibe-Trading](https://stockmarketstack.com/tools/vibe-trading.md) — A local finance research agent with 74 MCP tools, nine backtest engines and 14 brokers.

## FAQ

### What does it cost to run an open-source AI trading agent?

The code is free; the model calls and the data are not. AI Hedge Fund needs a Financial Datasets key, from $20 one-time for 1,000 requests, plus one model key, and caches every prompt so a re-run costs nothing. FinRobot needs an OpenAI key and a Financial Modeling Prep key. TradingAgents and Vibe-Trading read mostly keyless public data, so the model is the bill, and none of the four publishes a cost per run.

### Can any of these AI agents place real trades?

Only Vibe-Trading, which places live orders through eight brokers inside a mandate you set, with a kill switch, and which its own project calls experimental and unverified against a real account. AI Hedge Fund's only broker is an in-memory simulator, TradingAgents has no exchange or order code despite its README, and FinRobot connects to no broker. The copilots read; they do not trade.

### Can an individual subscribe to an AI research copilot like AlphaSense?

In practice, no. AlphaSense publishes no price, has no self-serve tier and sells annual seat contracts to firms, and Fintool, the challenger that sold the same way, was bought by Microsoft in April 2026 and shut down. The copilots a private investor can buy sit inside research products listed elsewhere, such as Fiscal.ai, or Quartr, whose mobile app includes its AI chat for free.

### Are the ratings these AI agents produce investment research?

No. They are generated output that nobody signs. TradingAgents closes on a five-tier rating that its own documentation says may differ between two runs of the same ticker and date, FinRobot's report reprints the sell-side rating and target it fetched from a data API, and an AI Hedge Fund persona is a system prompt. This catalogue describes what each one outputs and endorses none of it.