Connect SillyTavern to a DeepSeek API
SillyTavern's Chat Completion mode supports a custom OpenAI-compatible source. This page gives the exact values for this gateway's DeepSeek model, taken live from this gateway's own /v1/models endpoint.
1. Switch to Chat Completion
- Open the API Connections panel.
- In the API type drop-down, choose Chat Completion (not Text Completion).
- For Chat Completion Source, choose Custom (OpenAI-compatible).
2. Fill in the fields
| Field | Value |
|---|---|
| Custom endpoint / Base URL | https://api.privatedeskai.com/v1 |
| API key | YOUR_SAGG_KEY (from the account page) |
| Model | deepseek-ai/DeepSeek-V4-Flash-0731 (pick it from the dropdown once it loads, or type it in if SillyTavern shows a plain text field) |
SillyTavern's own documentation says the endpoint should end in /v1, and explicitly warns not to add /chat/completions yourself - SillyTavern appends the right path on its own.
3. Test the connection
Use SillyTavern's own Test Message button to confirm it connects and answers before starting a chat.
Model id and alias
The model id above, deepseek-ai/DeepSeek-V4-Flash-0731, and its short alias deepseek-v4-flash-fast, are both listed live at /v1/models - if SillyTavern's model dropdown does not show it, this API implements /v1/models, so the dropdown should list it once the base URL and key are set; if the dropdown is still empty, type the model id manually.
If something does not work
Double-check the base URL has no trailing /chat/completions and that the API key was pasted without extra spaces. For what request parameters this API accepts, see Compatibility; for current uptime and speed, see Status.
Source for the SillyTavern steps above: the official documentation at docs.sillytavern.app, page "Chat Completions" (section "Custom OpenAI-compatible endpoint"), checked 2026-11-14. The exact on-screen field labels are not stated on that page and may differ slightly by SillyTavern version.