All guides

SillyTavern

Easiest

The power-user standard for desktop roleplay and chat. Native custom endpoint, the cleanest setup.

The URL SillyTavern asks for Custom Endpoint (Base URL) https://api.vaelico.ai/v1

Steps

  1. Create your key in the app: open Account, then API keys, then New key (it starts with vk_).

  2. In SillyTavern, open the API Connections panel (the plug icon at the top).

  3. Set API to Chat Completion.

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

  5. Paste https://api.vaelico.ai/v1 into Custom Endpoint (Base URL). Do not add /chat/completions: SillyTavern appends it.

  6. Paste your vk_ key into Custom API Key and click Connect. If the key is right, the model list loads.

  7. Pick artemis from Available Models.

  8. Send a Test Message to confirm. The first message with a long imported history takes a while (the prompt is processed once); after that, the cache makes it fast.

Good to know

Full sampler passthrough: temperature, top_k, top_p, min_p, repetition penalty, presence and frequency penalty, seed, stop and logit_bias all reach the model, so your sampling preset applies as is.

Prefill works: "Start Reply With" and Continue send a trailing assistant message and Artemis continues it. The reply contains only the continuation; your prefill is not echoed back.

Different menus in your version of SillyTavern? Tell us: hello@vaelico.ai

Create my key (opens in a new tab) All guides