ZCode
Connect OpenGateway to ZCode as an OpenAI-compatible custom provider. Set its OpenAI base URL and API key in Model Settings, then add the UltraFast model IDs you want to use.
opengateway-zcode skill.Add OpenGateway as a custom provider#
1. Get an API key#
OpenGateway → API Keys → Create Key.
2. Open Model Settings#
In ZCode, click the current model name in the chat box to open the model selector, then choose Manage Models at the bottom. This opens Model Settings. The official ZCode Connect Models guide also documents this setup and its current UI labels.
3. Add the provider#
At the bottom of the provider list, click Add Provider. Give it a name,
such as OpenGateway. Set Endpoint URL to
https://apis.opengateway.ai/v1, choose Chat Completions (/chat/completions)
for API Format, and enter your key in API Key. Click Add Model and
add either or both IDs below. Save the provider, then turn on its enable
switch. ZCode's official guide documents adding an OpenAI-compatible endpoint
to a custom provider and entering the service's model IDs manually.
Choose Chat Completions (/chat/completions) rather than an Anthropic API
format. OpenGateway's OpenAI-compatible API uses the /v1 base URL.
4. Select and verify#
Choose the OpenGateway model from ZCode's model selector and send a short test prompt, such as "Reply with OK". ZCode's guide recommends verifying the connection with a short instruction.
Choose a model#
Add either or both of these OpenGateway UltraFast models in ZCode. Enter each ID exactly, including its creator prefix:
deepseek/deepseek-v4.1-flash-ultrafastz-ai/glm-5.3-flash-ultrafast
See UltraFast for more about these models.