> ## Documentation Index
> Fetch the complete documentation index at: https://veniceai-docs-revamp.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Jan AI

> Use Jan AI with Venice through an OpenAI-compatible remote model provider

[Jan](https://jan.ai/) is an open-source AI assistant for running local and remote models from one desktop app. Venice works in Jan through its OpenAI-compatible remote model provider support.

<CardGroup cols={3}>
  <Card title="Desktop Assistant" icon="desktop">
    Use Venice models inside Jan's chat interface
  </Card>

  <Card title="OpenAI Compatible" icon="plug">
    Use Venice's OpenAI-compatible chat API
  </Card>

  <Card title="Local + Remote" icon="layer-group">
    Keep local Jan models alongside Venice cloud models
  </Card>
</CardGroup>

***

## Prerequisites

* A Venice API key from [venice.ai/settings/api](https://venice.ai/settings/api)
* [Jan](https://jan.ai/) installed and running on your machine
* A Venice text model ID from the [model catalog](/models/text)

***

## Setup

<Steps>
  <Step title="Get your Venice API key">
    Generate or copy an API key from [venice.ai/settings/api](https://venice.ai/settings/api).
  </Step>

  <Step title="Open Jan's model provider settings">
    In Jan, open **Settings** and go to **Model Providers**.

    Choose Jan's OpenAI-compatible custom provider flow. Depending on your Jan version, this may appear as **OpenAI** with **Custom Endpoint** enabled, or as an **Any OpenAI Compatible API** provider.
  </Step>

  <Step title="Configure the Venice endpoint">
    Enter the Venice provider settings:

    | Field         | Value                          |
    | ------------- | ------------------------------ |
    | Provider name | `Venice AI`                    |
    | Base URL      | `https://api.venice.ai/api/v1` |
    | API key       | Your Venice API key            |
    | API type      | OpenAI-compatible              |

    Save or apply the provider settings before adding models.
  </Step>

  <Step title="Add Venice models">
    Add the Venice model IDs you want to use. Jan requires the model ID to match the provider exactly.

    Good starting options:

    * `zai-org-glm-5-1`
    * `kimi-k2-5`
    * `minimax-m25`

    You can add any supported text model from the [model catalog](/models/text).
  </Step>

  <Step title="Start a chat">
    Open a new or existing Jan chat, select one of your Venice models from the model selector, and send a test message.

    Jan will send the request to Venice using your API key and the OpenAI-compatible base URL.
  </Step>
</Steps>

***

## Verify the Setup

Send a short prompt from Jan:

```txt theme={"dark"}
Reply with one sentence confirming you are running through Venice.
```

If the provider is configured correctly, Jan should stream a response from the selected Venice model.

You can also verify your API key outside Jan with:

```bash theme={"dark"}
curl https://api.venice.ai/api/v1/models \
  -H "Authorization: Bearer $VENICE_API_KEY"
```

***

## Config Notes

* Use `https://api.venice.ai/api/v1` as the base URL. Do not include `/chat/completions`; Jan appends the endpoint path for chat requests.
* Model IDs must match Venice exactly. If a model does not appear in Jan's selector, add it manually using the model ID from the [model catalog](/models/text).
* Keep Venice configured as an OpenAI-compatible provider. Do not use Jan's native Anthropic provider for Venice Claude models.
* Remote Venice requests require an internet connection and consume Venice API credits.

<Info>
  Jan's local models remain available after you add Venice. You can switch between local and Venice models from the chat model selector.
</Info>

***

## Troubleshooting

<AccordionGroup>
  <Accordion title="401 or authentication error">
    Check that the API key is a Venice key from [venice.ai/settings/api](https://venice.ai/settings/api), with no extra spaces before or after it.
  </Accordion>

  <Accordion title="404 or model not found">
    Confirm that the model ID in Jan matches a Venice model ID exactly. Also check that the base URL is `https://api.venice.ai/api/v1`.
  </Accordion>

  <Accordion title="No Venice models in the selector">
    Add the model manually in Jan's provider settings. Use the exact model ID, such as `zai-org-glm-5-1`, and save the provider again.
  </Accordion>

  <Accordion title="Requests still go to another provider">
    Make sure the chat is using a model attached to your Venice provider. If Jan is using a built-in OpenAI or Anthropic model entry, switch to the Venice model you added.
  </Accordion>
</AccordionGroup>

***

## Resources

<CardGroup cols={2}>
  <Card title="Venice Text Models" icon="list" href="/models/text">
    Available Venice model IDs
  </Card>

  <Card title="Jan Remote Models" icon="book" href="https://www.jan.ai/docs/desktop/manage-models">
    Jan model provider documentation
  </Card>
</CardGroup>