跳转到主要内容
SillyTavern connects to Qhaigc using the Chat Completion API type with Custom (OpenAI-compatible) as the source. This routes all requests through Qhaigc’s OpenAI-compatible endpoint rather than a dedicated provider-specific channel.

​
Prerequisites

  • A running SillyTavern instance (desktop or self-hosted) that supports a custom Chat Completion source
  • A Qhaigc API key — get yours from the API Tokens page

​
Configure SillyTavern

1

Open the API configuration page

Open SillyTavern, go to Settings, and navigate to the API Connections (or API configuration) page.
2

Select Chat Completion as the API type

Set the API type to Chat Completion.
3

Set the Chat Completion source

Set Chat Completion Source to Custom (OpenAI-compatible).
4

Enter your Qhaigc credentials

Fill in the following fields:
FieldValue
Custom endpoint URLhttps://api.qhaigc.net/v1
API keyYour Qhaigc API key (starts with sk-)
Save the configuration.
5

Select a model

If SillyTavern can fetch the model list from /v1/models, choose a model from the dropdown.If the list is empty, type the model ID directly into the model field (for example, gpt-4o or claude-3-5-sonnet-20241022).
6

Start chatting

Reload the page or start a new chat and send a test message to confirm the connection works.

​
Verify the Connection

  • Send a test message to the bot.
  • A successful reply confirms that SillyTavern is connected to Qhaigc.
Enter only the base endpoint URL (https://api.qhaigc.net/v1) in the custom endpoint field. Do not append /chat/completions or any other path — SillyTavern appends the correct path automatically.

​
Troubleshooting

Model list is empty. Check whether Qhaigc’s /v1/models endpoint is reachable from your SillyTavern instance. If it is not, enter the model ID manually. Bot does not reply. Confirm that the endpoint URL ends at the base address level (/v1) and that your API key is correct.