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

# Bring your own model

> Connect your own model provider inside the Workspace.

## Your model, your credentials

EDN never supplies a model. In the Workspace, you connect the provider and model
you want to use through the settings opened from the chat status row.

## Providers

The settings support these providers:

* Anthropic, using an Anthropic API key.
* OpenAI, using an OpenAI API key.
* NVIDIA NIM, using an NVIDIA API key. The NVIDIA NIM free tier is available
  from build.nvidia.com at 40 requests per minute.
* Local or self-hosted, using an inference endpoint URL and an optional API key.

You then choose a model from the list of models available for the connected
provider.

## Inference settings

You can set the temperature, the maximum number of tokens, and an optional seed
for the response.

<Warning>
  Your provider key lives only in the settings for that session. It is not
  saved to your browser or persisted by the Console.
</Warning>