Lyra Core Chat

Chat Backend

Which model generates Lyra's replies. (Embeddings are set separately, via EMBED_BACKEND.)

Local — Ollama Free, private, runs on your home lab (LOCAL_MODEL) MI50 — local GPU Free, llama.cpp on the MI50 box (MI50_BASE_URL) Cloud — OpenAI Higher quality, costs money (CLOUD_MODEL)

Chat Model (Cloud)

Which OpenAI model answers on the Cloud backend. Tools (poker, equity, journaling) require Cloud.

Inner Voice (introspection)

Which model runs her reflections & thoughts (her dream loop). Dolphin is richer but shares the 3090 — switch to MI50 or Off before gaming.

Session Management

Manage your saved chat sessions:

Loading sessions...

Mode

Session

Actions

Settings

Chat Backend

Chat Model (Cloud)

Inner Voice (introspection)

Session Management