데이터베이스 섹션 · 12 레코드
벤치마크
How capability and hardware are actually measured — and which yardsticks are already saturated.
현장 노트
벤치마크는 이 업계가 점수를 매기는 방식이며, 생애주기는 언제나 같다: 도입, 정복, 은퇴. 각 기록에서 흥미로운 질문은 현재 선두가 누구냐가 아니라 — 테스트가 불가능에서 포화 상태까지 얼마나 빨리 도달했느냐다.
이 세 가지부터 시작하세요
ML
MLPerf
AI 하드웨어에서 유일하게 감사된 동등 비교 수치.
SB
SWE-bench Verified
실제 GitHub 이슈 — 코딩 에이전트 시대의 기준 척도.
AA
ARC-AGI
암기를 막도록 설계되었지만 뚫렸다. 그리고 더 단단하게 재구축되었다.
Hardware & systems benchmarks
2
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
10
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.