OpenCode / Hermes Agent

OpenAI-compatible coding agents such as OpenCode and Hermes Agent can route their requests through Qubax by setting a custom provider. Both read a base_urland api_key, after which every model call is billed against your Qubax credits.

Note: You will need a Qubax API key (starts with qbx_live_) and the base URL https://api.qubax.ai/v1.

OpenCode

OpenCode supports custom OpenAI-compatible providers in its config file. Add a Qubax provider entry pointing at the Qubax base URL.

JSON
{
  "providers": {
    "qubax": {
      "base_url": "https://api.qubax.ai/v1",
      "api_key": "qbx_live_your_api_key_here",
      "models": {
        "gpt-5": {},
        "claude-sonnet-4.5": {}
      }
    }
  },
  "default_provider": "qubax"
}

You can also pass the credentials via environment variables so they are never written to disk:

Shell
export OPENAI_BASE_URL="https://api.qubax.ai/v1"
export OPENAI_API_KEY="qbx_live_your_api_key_here"
Tip: List any Qubax model IDs you want to use in the modelsblock. Run curl https://api.qubax.ai/v1/models to see the full list available on your account.

Hermes Agent

Hermes Agent has official Qubax plugins. The provider plugin adds Qubax AI to the Hermes model picker with every Qubax text model. The media plugin uses Qubax for image generation, video generation, text-to-speech and speech-to-text. One key powers both.

Shell
hermes plugins install QubaxAi/hermes-qubax/provider
hermes plugins install QubaxAi/hermes-qubax/media
echo 'QUBAX_API_KEY=qbx_live_your_api_key_here' >> ~/.hermes/.env

# Chat: pick "Qubax AI" in the list, or start directly
hermes model
hermes --provider qubax -m glm-5.3-flash

To use Qubax for images, video and voice, run hermes tools or add this to your config:

YAML
# ~/.hermes/config.yaml
image_gen: {provider: qubax}
video_gen: {provider: qubax}
tts:       {provider: qubax}
stt:       {provider: qubax}
Tip: Model lists are read live from Qubax, featured models first, so new models show up without updating the plugin. Run /usage inside Hermes to see your balance. Source: github.com/QubaxAi/hermes-qubax.

Verifying the connection

After configuring, confirm the provider responds with a quick model list call:

Shell
curl https://api.qubax.ai/v1/models \
  -H "Authorization: Bearer $QUBAX_API_KEY"

If the call returns a JSON list of models, the provider is wired up correctly and your agent will route all requests through Qubax.

Ready to build?

380+ models through one OpenAI & Anthropic-compatible API, up to 99% below OpenRouter.

Get your API key →