Skip to main content
Ship Name Lab

Original research

Name Pair Benchmark

Twenty deliberately varied pairs run through the same deterministic engine used by Ship Name Lab. Inputs, thresholds, top results, and failures are public so changes can be checked instead of trusted.

Cases
20
Passing
20/20
Engine
2026.07.03-3
Reviewed
2026-07-03

What counts as a pass

Each case declares a minimum score and a maximum warning count before the engine runs. A passing result must produce a candidate using material from both inputs, meet that score floor, and stay within the warning limit. The benchmark does not claim that the first result is objectively the best nickname; human taste and pronunciation still decide that.

Common Latin-script cases use stricter thresholds. Diacritic and non-Latin cases intentionally permit one warning because the current readability model does not claim language-specific pronunciation knowledge. That limitation is visible rather than hidden.

Current run

Generated at request time from engine 2026.07.03-3.

Download machine-readable results
PairCategoryTop candidateScoreWarningsThresholdResult
Taylor + TravisSimilar opening consonants create several plausible boundaries.Common LatinTayvis99082; ≤0Pass
Alex + JordanTests a short first input against a longer second input.Common LatinAledan99082; ≤0Pass
Sofia + MateoBoth inputs contain open vowel sequences.Common LatinSofteo97082; ≤0Pass
Emma + NoahTests two compact, vowel-heavy names.Common LatinEmmoah99082; ≤0Pass
Chloe + MasonThe written and spoken boundaries may not agree.Common LatinMasloe97080; ≤0Pass
Daniel + OliviaTests two medium-length names with many possible cuts.Common LatinDanvia97082; ≤0Pass
Bo + JoTwo-letter inputs leave almost no room for source-preserving cuts.ShortBojo77075; ≤1Pass
Ann + LeeTests compact inputs and repeated-letter handling.ShortAnee94078; ≤1Pass
Kai + AvaBoth names are short and vowel-heavy.ShortKava94078; ≤1Pass
Elizabeth + JonathanTests whether the engine avoids simply joining two long names.LongJonabeth99082; ≤0Pass
Christopher + AlexandraA large cut space should still produce compact leaders.LongAlexopher92082; ≤0Pass
Aaliyah + MuhammadTests repeated letters and differing source lengths.LongMuhyah99080; ≤1Pass
Mary-Jane + O'ConnorSeparators are normalized before candidate construction.PunctuationMarnor97080; ≤0Pass
Anne-Marie + Jean-LucTests two compound names after punctuation normalization.PunctuationJeanarie97080; ≤0Pass
Zoë + ChloéLatin diacritics are retained and flagged for human pronunciation review.DiacriticsZoloé93175; ≤1Pass
José + MaríaTests source preservation when accented letters are present.DiacriticsMarosé99175; ≤1Pass
Łukasz + ZofiaThe non-ASCII join requires review by a relevant-language speaker.DiacriticsZofasz99075; ≤1Pass
小明 + 小红Character slicing is deterministic, but pronunciation is not scored.Non-Latin小明小红76165; ≤1Pass
はるか + れんKana inputs expose the boundary of the Latin readability heuristic.Non-Latinはるかれん76165; ≤1Pass
민준 + 서연Hangul output is generated but explicitly left for human language review.Non-Latin민준서연80165; ≤1Pass

Known limits

The engine scores visible character structure, not cultural meaning, gender, relationship suitability, trademark status, handle availability, or pronunciation in a specific language. Non-Latin results are candidates for human review, not linguistic validation. New engine versions may change rankings; the version and review date above make those changes auditable.