LLM Studio ACTIVE SERVICE

Unified LLM Studio Service

Universal prompt completion API routing transparently between OpenAI ChatGPT and Google Gemini with system persona instructions, file attachments, and web search grounding.

Universal Generation Endpoint

POST /v1/llm/generate

{
  "provider": "gemini",
  "model": "3.8 Flash",
  "prompt": "Explain the Voyager Restli 2.0 protocol architecture",
  "systemInstruction": "You are a senior systems architect.",
  "conversationMode": "ongoing",
  "continueConversation": true,
  "conversationId": "optional_conv_id",
  "newChat": false,
  "temporaryChat": true,
  "stream": false
}

Conversation Continuity Controls:

newChat: true — Forces a brand new chat thread every time (used by default in SocialMate).
continueConversation: true — Maintains the active conversation turn-by-turn (even in Incognito/Temporary mode).
conversationId: "..." — Directs message into a specific conversation thread.
temporaryChat: true (or incognito: true) — Enables ephemeral temporary chat with no history saved permanently to the provider account.

Supported Models

OpenAI: auto, gpt-4o, o1, o3-mini, research
Google Gemini: gemini-2.0-flash, gemini-1.5-pro, gemini-3.8-flash