Skip to main content
NextChat, also known as ChatGPT Next Web, can run as a web or desktop chat client. It supports custom OpenAI-compatible endpoints and is useful for personal or team chat entry points.

​
Values

FieldValue
Endpoint typeOpenAI / custom OpenAI
API endpointhttps://moxus.ai/v1
API Keysk-your-key
ModelCopy the exact model name from Models and pricing

​
UI setup

  1. Open NextChat settings.
  2. Find model provider or endpoint settings.
  3. Choose OpenAI.
  4. Set endpoint to https://moxus.ai/v1.
  5. Set API Key to sk-your-key.
  6. Save and choose or enter a model name.

​
Self-hosted environment variables

For self-hosted deployments, configure endpoint and key with environment variables: 
OPENAI_API_KEY=sk-your-key
BASE_URL=https://moxus.ai/v1
Variable names differ by version. If your deployment uses OPENAI_BASE_URL, API_BASE_URL, or another field, follow your version’s documentation.

​
Recommendations

  • Do not put an admin key in public frontend config.
  • Use separate API keys for each deployment, such as nextchat-prod and nextchat-test.
  • Add login or access control before exposing a self-hosted instance publicly.

​
Troubleshooting

Technically yes, but it is risky. Add access control in front of NextChat and set a quota limit on the key.
Restart the service and confirm your variable names match the current NextChat version. For Docker, confirm the container received the new values.
Enter the exact model name from Models and pricing manually.

​
Next steps