> ## 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.

# Kimi (Moonshot AI)

> Use your Moonshot AI API key with Rumus to run Kimi K2.5, K2 Thinking, and Moonshot V1.

Bring your own Moonshot AI key to use Kimi in Rumus — K2.5 with vision and tools, K2 Thinking for harder reasoning, and the Moonshot V1 family.

## Before you start

You need:

* A Moonshot AI account.
* An API key from the dashboard:
  * International: [platform.moonshot.ai](https://platform.moonshot.ai/) — base URL `https://api.moonshot.ai/v1`
  * China: [platform.moonshot.cn](https://platform.moonshot.cn/) — base URL `https://api.moonshot.cn/v1`
* Sufficient credit on your Moonshot account.

<Note>
  Keys from the international portal and the China portal are not interchangeable. Pick the one that matches the base URL you use.
</Note>

## Add Kimi 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 **Kimi**.
  </Step>

  <Step title="Paste your API key">
    Paste the key into **API Key**. It's stored encrypted in your local [vault](/docs/account/vault).
  </Step>

  <Step title="Base URL">
    Default is `https://api.moonshot.ai/v1` (international). If you're using a China-portal key, change it to `https://api.moonshot.cn/v1`.
  </Step>

  <Step title="Pick a model">
    Choose from the list, or toggle **Enter custom ID** to type a model ID manually.
  </Step>

  <Step title="Capabilities">
    Capability flags are pre-set for the built-in models. For a custom model, mirror what Moonshot's docs say it supports.
  </Step>

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

## Recommended models

| Model                           | ID                                                      | Good for                                                             |
| ------------------------------- | ------------------------------------------------------- | -------------------------------------------------------------------- |
| **Kimi K2.5**                   | `kimi-k2.5`                                             | Default daily driver — vision, tools, and thinking, 256K context     |
| **Kimi K2 Thinking**            | `kimi-k2-thinking`                                      | Longer thinking budget, no vision, 256K context                      |
| **Moonshot V1 128K / 32K / 8K** | `moonshot-v1-128k`, `moonshot-v1-32k`, `moonshot-v1-8k` | Earlier-generation general-purpose models in different context sizes |

For the full lineup, see the [Moonshot AI docs](https://platform.moonshot.ai/docs/).

## Tips

* **Pick K2.5 first.** It's the most capable everyday model — strong at tool use, supports vision, and handles long context.
* **K2 Thinking** drops vision but spends more on reasoning. Reach for it on tricky multi-step problems where extra deliberation helps.
* **Moonshot V1** is older. Useful if a workflow specifically depends on it; otherwise prefer K2.5.

## Troubleshooting

<AccordionGroup>
  <Accordion title="401 Unauthorized">
    The key is wrong, expired, or doesn't match the base URL. Confirm whether you're on the international or China portal and use the matching key + URL.
  </Accordion>

  <Accordion title="Model not in the list">
    Toggle **Enter custom ID** and paste the exact model ID (e.g. `kimi-k2.5`).
  </Accordion>

  <Accordion title="High latency on first response">
    Streaming first-token latency depends on region. Pick the portal that's closer to you.
  </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, OpenAI, Google, Z.AI, DeepSeek, Ollama, OpenAI-compatible.
  </Card>

  <Card title="Built-in models" icon="cube" href="/docs/models/built-in">
    Use Rumus's bundled Kimi without managing a key yourself.
  </Card>
</CardGroup>