Definition
Token minning is an emerging counterterm for systematically reducing token consumption while preserving an explicit threshold for useful outcome quality. Techniques include retrieving smaller context slices, stabilizing cacheable prefixes, shortening tool payloads, routing by task, bounding retries, replacing deterministic work with code, and stopping when additional inference no longer changes the decision.
The doubled n is intentional wordplay on tokenmaxxing and minimizing.
Where the term came from
The Tokenminning Manifesto presents itself as a founding document and defines the practice as minimizing token expenditure without sacrificing outcome quality. It is direct evidence of the intended spelling and meaning, but it is advocacy rather than independent proof of the claimed savings or environmental benefits. The techniques themselves predate the label: selective retrieval, caching, routing, bounded retries, deterministic tools, and stop rules are established ways to reduce unnecessary inference.
Why it is contested
The label is new, and its founding source is also advocating for its adoption. Independent evidence supports many of the underlying efficiency techniques, but it does not yet establish token minning as a distinct or standardized discipline. The spelling is also deliberate branding rather than conventional English. This glossary retains the term for the named movement while separating it from the broader, established goal of token efficiency.
Distinguish it from nearby terms
Token minning is not choosing the smallest token count at any cost. If compression removes needed evidence, pushes work into human review, or lowers acceptance and durability, total system efficiency can get worse. The target is fewer tokens for an equivalent or better accepted outcome.
Check your understanding
Hold the acceptance test constant, reduce one source of token demand, and measure whether tokens per accepted task improve without moving cost into retries, review, or later defects.