LobeChat with Nbility

LobeChat / LobeHub exposes OpenAI provider and proxy URL settings. Configure both the model service and the model selected in a conversation.

Set up the model serviceCopy link

1. Open Settings → Language Models / Model Services.
2. Choose OpenAI or a custom OpenAI-compatible provider.
3. Enter a dedicated key and https://api.nbility.ai/v1 as the API proxy URL.
4. Copy an exact ID from Models, then enable or add it to the provider’s list.
5. Save and select the corresponding model in the assistant or conversation.

Self-hosted instancesCopy link

Releases exposing OPENAI_PROXY_URL can use server-side variables:

OPENAI_API_KEY=YOUR_API_KEY
OPENAI_PROXY_URL=https://api.nbility.ai/v1

Restart the deployment and test. Server and user settings may override each other depending on release. Do not place private keys in public browser environment variables.

Verify and troubleshootCopy link

Test text first and inspect Usage logs. Knowledge bases, embeddings, images, and plugins may require separate models or services.

If the model list is empty, enter the ID manually or verify token access with the Models API. If only an existing conversation fails, check its saved model and capability settings.

ReferencesCopy link

Adapted and rewritten for Nbility from aicentos’s LobeChat guide, licensed under CC BY 4.0. Its account, plan, and proprietary service instructions do not apply here. See the client’s official documentation.