Skip to content
← All PDF guides
FIELD GUIDE 05 / OCTOBER 2026

The Inference Engineer’s Lexicon

Understand the term. Name the decision it changes.

A linked reference from GPU execution and memory traffic to distributed decoding and service metrics. Worked cases expose the distinctions that matter in inference reviews.

77 pages106 terms · 8 worked casesSearchable PDF
Sign in to check download access

The complete PDF requires active paid Premium or an eligible full-access grant. Referral access does not include downloads.

Cover of The Inference Engineer’s Lexicon
THE DECISIONS INSIDE

Explain an unfamiliar term through its scope, mechanism, units and a concrete serving consequence.

  1. 01Locate execution resources and memory traffic
  2. 02Distinguish occupancy, activity and throughput
  3. 03Read compiler, graph and profiler terminology
  4. 04Connect collectives to inference state
  5. 05Use latency and goodput without mixing denominators

GPU reference reviewed October 24, 2026. Inference adaptation October 25, 2026.

Opening workbook page from The Inference Engineer’s Lexicon. The adjacent topics describe the full edition.
Preview: the opening inference workbook page.
PUT IT TO WORK

Continue on the serving path.

ANOTHER VIEW OF THE SAME SYSTEM

Keep the next reference nearby.

Open Weights, Serving Decisions →

Accelerators for Token Serving →