> ## Documentation Index
> Fetch the complete documentation index at: https://www.rumus.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenAI

> Use your OpenAI API key with Rumus to run GPT and o-series models.

Bring your own OpenAI key to use GPT-5, GPT-4.1, the o-series reasoning models, and any other model your account has access to. Requests go directly from your machine to OpenAI.

## Before you start

You need:

* An OpenAI account with API access at [platform.openai.com](https://platform.openai.com/).
* An API key from **API keys** in your OpenAI dashboard.
* Sufficient credit or an active billing setup on the OpenAI side.

## Add OpenAI in Rumus

<Steps>
  <Step title="Open the model settings">
    Go to **Settings → AI → Models** and click **Add Model**.
  </Step>

  <Step title="Pick the provider">
    Set **Provider** to **OpenAI**.
  </Step>

  <Step title="Paste your API key">
    Paste your `sk-...` key into **API Key**. The key is masked in the UI and stored encrypted in your local [vault](/docs/account/vault).
  </Step>

  <Step title="(Optional) Custom base URL">
    Leave **Base URL** blank to use `https://api.openai.com`. Override only if you're using an Azure deployment fronted by an OpenAI-compatible URL or a corporate proxy.

    <Tip>
      For Azure OpenAI specifically, the [OpenAI-compatible](/docs/models/openai-compatible) provider gives you finer control over headers and query parameters.
    </Tip>
  </Step>

  <Step title="Pick a model">
    Choose from the list (GPT-5, GPT-5 Mini, GPT-4.1, GPT-4o, o4-mini, o3, o1, etc.) or toggle **Enter custom ID** to type a model ID manually.
  </Step>

  <Step title="Capabilities">
    On the **Capabilities** tab:

    * **Tool Calling** — enable for agentic features (most GPT and o-series models support it).
    * **Vision** — enable for GPT-4o, GPT-4.1, and other multimodal models.
    * **Prompt Cache** — OpenAI's automatic prompt caching is supported on supported models.
  </Step>

  <Step title="Save">
    The model appears in the picker under **Custom Models**.
  </Step>
</Steps>

## Recommended models

| Model                         | Good for                                                      |
| ----------------------------- | ------------------------------------------------------------- |
| **GPT-5**                     | Heavy multi-step work, ambiguous prompts, hard reasoning      |
| **GPT-5 Mini / GPT-4.1 Mini** | Default daily driver — strong quality, lower cost             |
| **GPT-4o**                    | Vision-heavy tasks (screenshots, diagrams)                    |
| **o3 / o4-mini**              | Math, code reasoning, planning where extra "think time" helps |

For the latest model lineup and pricing, see the [OpenAI models page](https://platform.openai.com/docs/models).

## Tips

* **Reasoning models** (o-series) ignore `temperature` and use a thinking budget instead. Rumus handles the parameter difference automatically.
* **Prompt caching** is automatic on supported OpenAI models — leaving the capability flag on lets Rumus account for cache reads correctly.
* **Project-scoped keys** work fine. Create a separate key per project to keep usage tracked.

## Troubleshooting

<AccordionGroup>
  <Accordion title="401 Incorrect API key provided">
    The key is missing, expired, or revoked. Recreate it in the OpenAI dashboard.
  </Accordion>

  <Accordion title="403 / Model not accessible">
    Some models require a verified org or higher tier. Check **Limits** on the OpenAI dashboard.
  </Accordion>

  <Accordion title="429 Insufficient quota">
    Add credit to your OpenAI account or attach a payment method.
  </Accordion>

  <Accordion title="Model not in the list">
    Toggle **Enter custom ID** and paste the exact model ID (e.g. `gpt-5-2025-11-04`). The list ships with common defaults but not every snapshot.
  </Accordion>
</AccordionGroup>

<Note>
  Hit a snag we didn't cover? Ask in the [Rumus community](https://www.rumus.ai/community).
</Note>

## Next steps

<CardGroup cols={2}>
  <Card title="Other providers" icon="grid" href="/docs/models/overview">
    Anthropic, Google, Z.AI, DeepSeek, Kimi, Ollama, OpenAI-compatible.
  </Card>

  <Card title="OpenAI-compatible" icon="plug" href="/docs/models/openai-compatible">
    Azure OpenAI, OpenRouter, vLLM, and friends.
  </Card>
</CardGroup>