> For the complete documentation index, see [llms.txt](https://thegrid.ai/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://thegrid.ai/docs/start-here/introduction.md).

# Introduction

## The Grid is a Spot Market for AI Inference

You pick an instrument. For most workloads, that means picking a task type and a quality tier. We route your request to a qualifying supplier at the lowest live price for that combination.

Task types describe the workload:

* **Text** for general-purpose chat and reasoning.
* **Code** for code generation, editing, and review.
* **Agent** for tool use, function calling, and multi-step workflows.

Quality tiers describe how strong the underlying model has to be:

* **Standard** for everyday tasks where speed and cost matter most.
* **Prime** for production workloads that need consistent, capable output.
* **Max** for the hardest tasks where you want frontier-class quality.

Each task type and tier maps to an **instrument**, a published spec like `text-prime` or `code-max`. Suppliers compete on every order. Any model that drifts below spec is removed from the eligible set, so you get the price of competition without giving up the quality bar.

The Grid also offers lab latest markets such as `gpt-sol-latest` and `claude-opus-latest`. Use these when you want continuity with a model family or lab while still consuming it through a Grid instrument instead of a fixed upstream model endpoint.

{% hint style="info" %}
We're compatible with both OpenAI Chat Completions and Anthropic Messages formats. See the [API reference](/docs/api-reference/api-reference.md) for base URLs, auth headers, and request shape.
{% endhint %}

## What this means for you

* **Lower price for the quality you need.** Suppliers bid against each other on your order. Pricing tracks live supply and demand, not a published rate card.
* **One API, no SDK rewrite.** Keep your OpenAI or Anthropic client. Swap in an instrument string as the model name and ship.
* **No vendor lock-in.** You aren't tied to one model house. If a better-priced model qualifies, we route to it without you changing code.
* **Buying isn't a one-way decision.** Users who enable [Advanced Mode](/docs/start-here/auto-mode-vs-advanced-mode.md) can opt to hold their purchased tokens in their Trading account before consuming them. Tokens held in your Trading account can be resold on the order book, which gives you room to experiment with how much you buy. See [Reselling tokens](/docs/concepts/reselling.md).

## Where to go next

* [Quickstart](/docs/start-here/quickstart.md): make your first call in under five minutes.
* [Choose an instrument](/docs/start-here/choose-an-instrument.md): pick the task type, quality tier, or lab latest market for your workload.
* [Current instruments](/docs/instrument-specifications/current-instruments.md): see the live specs and what each one is benchmarked against.
* [Benefits](/docs/benefits/benefits.md): what you gain by consuming inference at market prices.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://thegrid.ai/docs/start-here/introduction.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.