AI Foundationspredict · compress · act

concepts → Mechanism

Mechanism

Inference

timeless5 connectionsdraft

running a trained model to get an answer out

Where it sits

The prism has six jobs across and eight layers down. Its primary cell isinfer × L5. Hatched cells cannot exist — a GPU does not learn, an institution does not infer.

What must come first — and what it unlocks

Left to right is reading order, derived from the prerequisite_of edges. Nothing here is hand-ordered: the diagram is the graph.

nothing comes first — this is a starting pointInferenceAgent

Unlocks: Agent

What kind of thing it is — and what it is made of

Examples of it: vLLM.

It is made of: Batching, Speculative decoding, Local serving.

Where to read it

The chapter that introduces it, and any chapter that uses it again.

38How Tokens Get Servedact 8 · The Agent Infrastructure

Where it comes from

Every connection

All 5 edges touching this node, grouped by relation family — the sections above are highlights from this list. Colours match the relation families inthe atlas.

Structure · 4
has as a partBatchingMechanism
has as an instancevLLMSystem
has as a partSpeculative decodingAlgorithm
has as a partLocal servingSystem
Order · 1
must come beforeAgentSystem

This page is a projection of one node in src/data/concepts.ts. It has no prose file of its own — 230 declared edges produce all 236 of these pages. Edit an edge and both endpoints change.