> For the complete documentation index, see [llms.txt](https://docs.zerve.ai/zerve-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/models-and-effort.md).

# Models and effort

How Zerve chooses which model the agent uses, and how to change a model's reasoning effort.

Zerve's agent can run on models from several providers. Your organization's admins decide which models are available, and you can choose between them.

For the full list of models, see [Supported models](/zerve-docs/agents-in-the-workspace/supported-models.md).

## How the model is chosen

| If…                       | The agent uses…                                                                                                  |
| ------------------------- | ---------------------------------------------------------------------------------------------------------------- |
| **You've chosen a model** | That model, as long as it's one your organization has enabled                                                    |
| **You haven't**           | Zerve's default for the kind of work, such as a stronger model for coding and planning and a faster one for chat |

For example, with Anthropic, coding and planning use **Claude Sonnet 4.6** by default, and chat uses **Claude Haiku 4.5**. See [Default models](/zerve-docs/agents-in-the-workspace/supported-models.md#default-models).

You can only choose models your organization has enabled. To use another model, ask an admin to enable it in [Model settings](/zerve-docs/administration/model-settings.md).

## Reasoning effort

Each model has a **reasoning effort**: **Low**, **Medium** or **High**. It sets how much the model thinks before it answers.

| Effort     | Best for                                                     | Trade-off                                                        |
| ---------- | ------------------------------------------------------------ | ---------------------------------------------------------------- |
| **Low**    | Quick questions and simple tasks                             | Fastest, and uses the fewest credits                             |
| **Medium** | Most work                                                    | A balance of quality, speed and cost                             |
| **High**   | Hard, multi-step problems, such as debugging a complex model | Best results on hard problems, but slower, and uses more credits |

Every model has a default effort, shown next to it in the model list. You can change the effort, and other options such as speed, for each model in your own settings.

Higher effort means the model produces more tokens while it reasons, so it costs more. See [How credits are used](/zerve-docs/administration/how-credits-are-used.md).

## Models for superagents

[Superagents](/zerve-docs/superagents/superagents.md) have their own model settings: one model for the superagent itself, and three for its workers, called **Small effort**, **Medium effort** and **Large effort**. The superagent picks one of the three for each task. See [Superagent settings](/zerve-docs/superagents/settings.md#model-and-effort).

{% hint style="info" %}
A superagent's **Small, Medium and Large effort** settings choose **which model** a worker uses. A model's **Low, Medium and High** effort controls **how much it reasons**. They're separate settings.
{% endhint %}

In superagent conversations, you don't choose a model; the superagent uses its settings.

## Your own API keys

Organizations can use their own accounts with model providers instead of Zerve's. See [Bring your own API keys](/zerve-docs/administration/bring-your-own-api-keys.md).


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/models-and-effort.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.