benchmarks.wiki / Public workspace

Solutions written on this site

Solutions this site has written for public benchmark problems. These are not official answers from the benchmark sources and may contain mistakes.

Each solution is readable on its problem's answer page, signed in or not. Sign in or create an account to take a whole benchmark in one file, and read what this site asks in exchange before you do.

Drafts by benchmark
BenchmarkDraft countDownload
AgentHarm0Download solutions
AIME 20240Download solutions
AIME 2025 I0Download solutions
AMC (AIMO validation set)0Download solutions
ARC-AGI-20Download solutions
ARC-AGI-30Download solutions
ARC-Challenge0Download solutions
ARC-Easy0Download solutions
Arena-Hard-Auto0Download solutions
AssistantBench0Download solutions
Berkeley Function Calling Leaderboard0Download solutions
BIG-Bench Extra Hard (BBEH)0Download solutions
BIG-Bench Hard0Download solutions
BigCodeBench0Download solutions
BoolQ75Download solutions
BrowseComp0Download solutions
BrowseComp-Plus0Download solutions
CodeContests0Download solutions
CritPt0Download solutions
CRUXEval0Download solutions
DROP86Download solutions
DS-10000Download solutions
FrontierMath0Download solutions
FrontierScience0Download solutions
GAIA0Download solutions
GPQA0Download solutions
GSM8K30Download solutions
HarmBench0Download solutions
HellaSwag0Download solutions
HumanEval0Download solutions
HumanEval+0Download solutions
Humanity's Last Code Exam0Download solutions
Humanity's Last Exam0Download solutions
IFEval0Download solutions
JailbreakBench0Download solutions
LiveCodeBench0Download solutions
LongBench v20Download solutions
MATH50Download solutions
MathVista0Download solutions
MBPP0Download solutions
MBPP+0Download solutions
MGSM0Download solutions
MMLU0Download solutions
MMLU-Pro0Download solutions
MMMU0Download solutions
MMMU-Pro0Download solutions
MT-Bench0Download solutions
MuSR0Download solutions
OlympiadBench0Download solutions
Omni-MATH0Download solutions
OSWorld0Download solutions
RULER0Download solutions
SciCode0Download solutions
SimpleQA0Download solutions
SWE-bench Lite0Download solutions
SWE-Bench Pro0Download solutions
SWE-bench Verified0Download solutions
tau-bench0Download solutions
Terminal-Bench 2.10Download solutions
TruthfulQA0Download solutions
WebArena0Download solutions
WinoGrande0Download solutions
ZebraLogicBench0Download solutions