Benchmarks | AI Infrastructure Database

Catatan lapangan

Benchmark adalah cara industri ini mengukur pencapaian, dan siklus hidupnya selalu sama: diperkenalkan, ditaklukkan, lalu pensiun. Pertanyaan menarik pada setiap entri bukan siapa pemimpin saat ini — melainkan seberapa cepat sebuah tes bergeser dari tampak mustahil menjadi jenuh.

Mulai dengan tiga ini

ML MLPerf Satu-satunya angka yang diaudit secara apples-to-apples dalam hardware AI. SB SWE-bench Verified Issue GitHub nyata — tolok ukur era agen pengodean. AA ARC-AGI Dirancang untuk menahan hafalan; tetap jebol. Lalu dibangun ulang lebih kuat.

Hardware & systems benchmarks

ML MLPerf The industry's referee for AI hardware — training and inference rounds. TO TOP500 The supercomputer league table since 1993 — now measuring the public sector's slice.

Model capability benchmarks

SB SWE-bench Verified Real GitHub issues, real repos — the coding-agent standard. A( AIME (Competition Math) Olympiad-track problems — where reasoning models showed their jump. AA ARC-AGI Chollet's abstraction puzzles — designed to resist everything but intelligence. FR FrontierMath Research-grade mathematics — built unsolvable, already yielding. GD GPQA Diamond Google-proof PhD science questions — the reasoning era's exam. H& HumanEval & GSM8K The retired classics of coding and grade-school math. HL Humanity's Last Exam 2,500 expert questions with a deliberately final name. L( LMArena (Chatbot Arena) Millions of blind human votes — preference as a leaderboard. M& MMLU & MMLU-Pro The default knowledge exam of 2020–24 — now saturated. MT MTEB The embedding leaderboard behind every retrieval stack.