Expose halo's [fast] MoE preset through the LiteLLM gateway and make it the rag CLI's default chat model (overridable via RAG_CHAT_MODEL), so query synthesis is quicker than the larger coder model. |
||
|---|---|---|
| .. | ||
| acme.nix | ||
| backup.nix | ||
| default.nix | ||
| fileserver.nix | ||
| firefly.nix | ||
| hardware-configuration.nix | ||
| litellm.nix | ||
| mail.nix | ||
| network.nix | ||
| nginx.nix | ||
| opencode.nix | ||
| openwebui.nix | ||
| qdrant.nix | ||
| searx.nix | ||
| uptime-kuma.nix | ||
| wyoming.nix | ||