Loading…
Runs a templated prompt against a configured model provider with token limits, retries, and redaction hooks for PII.
CHAT_LLM
Upstream peers: major → domain → VM type.
This node: Chat → Integration · AI / LLM / voice
Peers: 9 majors · 38 domains · 144 VM types
Downstream peers: major → domain → VM type.
This node: Chat → Integration · AI / LLM / voice
Peers: 6 majors · 31 domains · 145 VM types
Services, APIs, or integrations this node uses outside the graph VM.
Primary chat-completions compatible API.
https://api.inference.acme.dev/v1
Link status: connected
POST token counts for billing pipeline.
Link status: connected
Graphs that contain at least one instance of this node type (reverse dependency).
No graphs use this node yet.
This definition is marked beta. Runtime instances inherit health from the executor pool, published semver, and feature flags for your tenant.
This is a public system/built-in node and cannot be edited directly. To customize it, fork it or create a private overlay, then edit your own copy.
scope: public