STOBE

Dwemer Dynamics Home Overview Installation Configuration Roleplay Settings Kenshi Settings Large Language ModelsLocal LLMs Text-to-Speech Speech-to-Text Modders Guide FAQ Remote Hosting Guide Linux Setup Guide Log Files

⬇️Download 🎥Video Guide DiscordDiscord YouTubeYouTube PatreonPatreon 💻Server Code 💻Mod Code

STOBE Large Language Model Services

StobeServer can use several hosted and local AI services. OpenRouter is the simplest place to start, and profiles let you keep four response choices ready at once.

Response Modes

Each profile has four response slots. Switch between them from the in-game Chat Box without rebuilding the profile.

  • Standard: The normal choice for everyday dialogue. New profiles use DeepSeek V4 Flash.
  • Fast: Prioritizes quicker replies. New profiles use Gemini 2.5 Flash Lite.
  • Powerful: Use when a scene needs more careful reasoning. New profiles use GLM 5.2.
  • Experimental: A spare slot for trying another model. New profiles use DeepSeek V4 Pro.

QuickStart fills these defaults for new installs. Existing profile choices are not replaced during updates.

Services

OpenRouter

OpenRouter is the easiest general starting point for most STOBE users. It gives you one account that can access many different AI models, so you can try options without constantly rebuilding your setup.

OpenRouter

OpenAI

OpenAI connects STOBE straight to GPT models. It is a simple choice if you already use OpenAI and want a more direct setup instead of browsing lots of model options.

OpenAI

Google

Google uses Gemini models for your STOBE conversations. It is another mainstream cloud option if you prefer Google's AI tools or want to compare its style against OpenAI and OpenRouter.

Google AI Studio

Groq

Groq is mainly about speed. It can make STOBE feel more responsive because replies tend to come back quickly, which some players prefer for fast back-and-forth dialogue.

Groq

Nano-GPT

Nano-GPT is another hosted service that gives you access to AI models without running them on your own PC. Think of it as an alternative provider if you want more choice outside the biggest platforms.

Nano-GPT

Player2

Player2 lets STOBE use the Player2 app running on your computer. Choose Player2 in QuickStart if that is the service you want for all response modes.

Player2 Local Endpoint

Use Your Own LLM Provider

OpenRouter is optional. STOBE can connect to other hosted or local services through a Custom connector.

Before You Start

Get these details from your provider:

  • An API key, unless the service runs locally without one.
  • The complete chat-completions URL.
  • The exact model name the provider expects.

Create the Connector

  1. Open API Keys. Save the provider's key in a preset entry, or add a named custom key.
  2. Open Large Language Models, select New Connector, then choose Custom.
  3. Enter a clear Name, the complete URL, and the provider's exact Model name.
  4. Choose the matching Driver. Common choices include OpenAI JSON for OpenAI-compatible services, Google OpenAI JSON for Gemini, and Groq JSON for Groq.
  5. Select the API key you saved.
  6. Leave Reasoning Model Fix, Enforce JSON, JSON Schema, Prefill JSON, and Disable Streaming unchanged unless the provider specifically requires them.
  7. Click Save, then Test. Do not assign the connector until the test succeeds.

Example: Google Gemini

Gemini is one example of a provider you can add manually:

  • API key: Save your Google AI Studio key under the Google preset or a named custom key.
  • URL: https://generativelanguage.googleapis.com/v1beta/openai/chat/completions
  • Driver: Google OpenAI JSON
  • Model: Use the exact current model name shown in Google AI Studio.

The built-in Google connector is simpler for most users. Use Custom when you want manual control over the endpoint and model. See Google's OpenAI compatibility guide for its current endpoint and model guidance.

Use It in STOBE

Open Profiles and select the connector for Standard, Fast, Powerful, Experimental, Diary, Autochat, Memory, Background Life, Dynamic Profile, or Relationship. You can then switch the active conversation mode from the in-game Chat Box.

If the Test Fails

  • 401 or 403: Check the selected API key and the provider account.
  • 404: Check the full URL and model name.
  • No usable response: Make sure the selected driver matches the provider's response format.
  • Invalid JSON: Try a model that handles structured responses well. Only enable the advanced JSON options when the provider supports them.

Local Models

Use the launcher's LLM Studio component to run LM Studio inside DwemerDistro. Follow the Local LLM Guide to download, load, test, and connect a model.

STOBE also has guided local setup in Quickstart: choose the provider and endpoint, use Probe models, then Send test message. Review Apply to before selecting Apply to dialogue; it changes the selected Default Profile, Player Faction profile, or both. Use the guide's manual connector steps to test another profile first.