# Forward Model

- [Forward Model docs](https://docs.forwardmodel.ai/index.md): Forward Model is an AI research lab for the financial markets.
- [Quickstart](https://docs.forwardmodel.ai/quickstart.md): Get today's scores in your own code in three steps.
- [For coding agents](https://docs.forwardmodel.ai/for-agents.md): Everything a coding agent needs to use the Forward Model scores API, on one page.
- [What a score means](https://docs.forwardmodel.ai/what-a-score-means.md): A score is our AI agent's call on one stock, on one day: how it expects that stock to do against the other scored stocks over the next 21 trading days.
- [The scores API](https://docs.forwardmodel.ai/scores-api.md): One endpoint returns the published scores for the stocks you name.
- [Claude and ChatGPT](https://docs.forwardmodel.ai/claude-and-chatgpt.md): Forward Model works as a connector in AI assistants that support the Model Context Protocol (MCP).
- [Keys and sign-in](https://docs.forwardmodel.ai/authentication.md): Sign in at console.forwardmodel.ai with Google or with a code sent to your email.
- [Get scores](https://docs.forwardmodel.ai/api-reference/get-scores.md): The published scores for the stocks you name. Each stock counts once per Pacific day toward the plan, whichever day's score you get; asking again that day is free. At the limit the whole call is refused (limit_reached) and nothing counts.
- [Plans and limits](https://docs.forwardmodel.ai/plans-and-limits.md): Unlimited chat is subject to fair use.
- [Errors](https://docs.forwardmodel.ai/errors.md): Every error has the same shape: a `type` you can act on, a `message` in plain words, and the `request_id`.
- [Data and privacy](https://docs.forwardmodel.ai/data-and-privacy.md): We keep only what counting and support need, for each call:

## OpenAPI Specs

- [openapi](/openapi.json)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.