Sources

In use

SourceRoleLicenseResultsEveryUpdatedState
Epoch AI
Epoch AI
Evaluator CC BY 4.0 3,777 6h 2026-11-11 Healthy
MathArena
SRI Lab, ETH Zurich
Evaluator None stated 37 6h 2026-11-11 Healthy
3 unmatched
Vals AI
Vals AI
Evaluator None stated 1,516 6h 2026-11-11 Healthy
26 unmatched
models.dev
models.dev (SST)
Metadata MIT 0 6h 2026-11-11 Healthy
ARC Prize
ARC Prize Foundation
Original Internal use 633 6h 2026-11-11 Healthy
21 unmatched
BFCL
UC Berkeley Gorilla team
Original Apache 2.0 80 24h 2026-11-11 Healthy
29 unmatched
EQ-Bench
Sam Paech
Original None stated 262 24h 2026-11-11 Healthy
44 unmatched
Kagi
Kagi
Original None stated 104 24h 2026-11-11 Healthy
37 unmatched
LMArena
LMArena
Original CC BY 4.0 561 12h 2026-11-11 Healthy
148 unmatched
Lech Mazur
Lech Mazur
Original None stated 245 6h 2026-11-11 Healthy
9 unmatched
LiveBench
LiveBench (Abacus.AI and collaborators)
Original None stated 224 6h 2026-11-11 Healthy
2 unmatched
METR
METR
Original None stated 24 24h 2026-11-11 Healthy
1 unmatched
OSWorld-Verified
XLANG Lab, University of Hong Kong
Original None stated 27 24h 2026-11-11 Healthy
26 unmatched
SWE-rebench
Nebius
Original None stated 13 24h 2026-11-11 Healthy
SimpleBench
SimpleBench team (AI Explained)
Original None stated 101 6h 2026-11-11 Healthy
5 unmatched
Terminal-Bench
Stanford / Laude Institute (Harbor)
Original Apache 2.0 35 6h 2026-11-11 Healthy
tau2-bench
Sierra Research
Original MIT 51 24h 2026-11-11 Healthy

Reviewed, not used

SourceStatusWhy
SWE-bench boardRejectedStale; repeats Epoch's copy
LMArena Agent ArenaCandidateScore scale not documented
OpenCompass, MMMU, OCRBenchCandidateNo data license or stale
LiteLLM price listCandidatePrice cross-check only
Scale SEAL, HLE boardRejectedInternal use only; Epoch has HLE
Artificial AnalysisRejectedNeeds an API key
OpenRouterRejectedTerms forbid copying
Open LLM LeaderboardRejectedFrozen since March 2025
LLM-statsRejectedVendor numbers; API blocked
BenchLeader, AGI RankerRejectedWould count results twice
Vendor model cardsRejectedSelf-reported, not checked
LiveCodeBench, BigCodeBench, EvalPlusRejectedStale or saturated
Design ArenaRejectedTerms forbid crawlers
YuppRejectedBehind an anti-bot wall
Fiction.LiveBenchRejectedOn hiatus; image only
LiveCodeBench ProRejectedSite is gone
Aider polyglot (direct)RejectedStale; Epoch has a copy

Recent updates

StartedSourceOutcomeRowsNewChangedUnmatched
2026-11-11 15:33:24 UTCMETROk26001
2026-11-11 10:09:55 UTCVals AIOk1,6053085
2026-11-11 10:09:55 UTCTerminal-BenchUnchanged35000
2026-11-11 10:09:55 UTCSimpleBenchUnchanged111000
2026-11-11 10:09:49 UTCMathArenaUnchanged40000
2026-11-11 10:09:45 UTCLiveBenchUnchanged238000
2026-11-11 10:09:42 UTCLech MazurUnchanged260000
2026-11-11 10:09:41 UTCEpoch AIOk4,6093700
2026-11-11 10:09:40 UTCARC PrizeUnchanged714000
2026-11-11 10:09:38 UTCmodels.devOk580000
2026-11-11 03:52:44 UTCVals AIOk1,60510085
2026-11-11 03:52:44 UTCTerminal-BenchOk35000
2026-11-11 03:52:44 UTCtau2-benchOk51000
2026-11-11 03:52:44 UTCSWE-rebenchOk13100
2026-11-11 03:52:44 UTCSimpleBenchOk111405
2026-11-11 03:52:44 UTCOSWorld-VerifiedOk930065
2026-11-11 03:52:44 UTCMETROk26001
2026-11-11 03:52:44 UTCMathArenaOk40003
2026-11-11 03:52:44 UTCLMArenaOk74260168
2026-11-11 03:52:44 UTCLiveBenchOk2380014