Skip to main content
A custom endpoint in librechat.yaml puts NagaAI’s chat models into your own LibreChat, and the model list updates itself. LibreChat sends custom endpoints Chat Completions and has no Responses option for them. The entry follows the LibreChat custom endpoint reference. NagaAI also runs a hosted LibreChat at chat.naga.ac.

​
Before you start

You need a NagaAI API key and a self-hosted LibreChat.

​
Connect LibreChat

1

Add the endpoint

Add NagaAI to librechat.yaml:
2

Provide the key

Set NAGA_API_KEY in LibreChat’s .env. To let each user enter their own key, use apiKey: "user_provided" instead.
3

Check the connection

Start LibreChat with the new config, open a chat and pick NagaAI. The model list comes from /v1/models.

​
Endpoint settings

With fetch: true, LibreChat loads the model list from /v1/models, so new models appear without editing the file. LibreChat sends headers with that request, and x-naga-supported-endpoint: chat.completions limits the list to chat models. Without the header the list includes image, audio and embedding models. default is the list LibreChat shows when the fetch fails. reasoningFormat: reasoning_effort makes LibreChat send reasoning_effort, the field NagaAI reads.

​
Troubleshooting

​
401 invalid_api_key

The key is missing or wrong. Check NAGA_API_KEY in LibreChat’s .env, or the key a user entered, and use an API key rather than a provisioning key.

​
Image and audio models in the list

The x-naga-supported-endpoint header is missing from the endpoint’s headers. Add it as shown in the first step.