LoomCycle Chat Model icon

LoomCycle Chat Model

Route the AI Agent's LLM calls through loomcycle's gateway (provider routing, auth, retry, per-user quotas)

LoomCycle Chat Model

AI-generated

Summary

Routes an AI Agent’s chat model requests through the loomcycle gateway, which can select a provider and model using the configured resolver. Connect this node to an AI Agent’s Chat Model input; it is not a standalone workflow step.

Inputs

  • Provider — Optionally pin a provider configured in the deployment’s loomcycle.yaml. Leave blank to let the resolver choose.
  • Model — Optionally pin a model ID supported by the selected provider. Leave blank to let the resolver choose.
  • Tier — Optional operator-defined dispatch tier; blank uses the resolver’s default.
  • User ID — Optional per-user quota and audit-log scope. If blank, uses the credential’s Default User ID; if that is also blank, requests use anonymous operator-bearer scope.
  • User Tier — Optional per-user tier overlay. If set, it takes precedence over Tier; otherwise the node falls back to Tier or the credential default.
  • Max Tokens — Maximum generated tokens per call; defaults to 4096.
  • Temperature — Sampling temperature; defaults to -1, which leaves the provider’s default in effect.
  • Streaming — Enable token-by-token streaming; defaults to true. Disable it for consumers that require a single non-streaming response.

Output shape

An n8n AI Language Model connection for an AI Agent, supplied through the node’s model response.

Requires LoomCycle API credentials. Supports tool calling when used with an AI Agent that provides tools. Chat responses include text and can include tool calls; the model exposes token usage and provider/model metadata through LangChain. The node does not return regular workflow items.

Links

Discussion