Use Chatbox AI with llmrelay

Setup

1. 1 Open Chatbox and click the settings button in the sidebar.
2. 2 Open the Model Provider section and click Add at the bottom of the provider list.
3. 3 Enter a provider name and choose the OpenAI API compatible type.
4. 4 Fill API Key with your relay key and API Host with https://api.llmrelay.dev/v1; leave API path empty so it defaults to /v1/chat/completions.
5. 5 Add at least one model (e.g. gpt-5.6-luna) and tick its capability flags, then save.
6. 6 Click Check and confirm it reports a successful connection.

Config

# Chatbox: Settings -> Model Provider -> Add
# Name:     LLM Relay
# API type: OpenAI API compatible
# API Key:  ***
# API Host: https://api.llmrelay.dev/v1
# API path: (leave empty -> defaults to /v1/chat/completions)
# Models:   gpt-5.6-luna, claude-sonnet-5

Smoke test

Copy, paste, run. If you get JSON back with a message, you're good.

Press the Check button in the provider settings; it should show connection successful. Then start a new conversation on that model.

Gotchas

• !Models added without capability flags are treated as plain-text models, so vision or reasoning features stay off until you tick them.
• !API path defaults to /v1/chat/completions, so a host that already ends in /v1/chat/completions will produce a doubled path.
• !The docs' provider types are protocol families (OpenAI API compatible, Google Gemini compatible); there is no Anthropic Messages type, so Claude models must go through the relay's OpenAI-compatible surface.

Related integrations