Kimi K3

Kimi K3 ranks first on 5 of the 214 benchmarks it has scores on.

Moonshot AI Released 16 Jul 2026 Family Kimi K3 The lab's model card

238

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Moonshot AI Jul 2026

Kimi K3

Open weights Flagship

158.3 #20 of 271

Epoch Capabilities Index

MMMU-Pro 81.6% #7 /72
DeepSWE 68.5% #8 /28
SciCode 58.7% #4 /47

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

Benchmark	Score	Rank	Source
Creative Writing v3	81.8	#4 of 120	Official leaderboard
LMArena (Elo)	1,488	#16 of 333	Official leaderboard
AA-LCR (Artificial Analysis Long Context Reasoning)	74.7%	#3 of 31	Model card
Extended NYT Connections	93.6%	#11 of 97	Official leaderboard
EQ-Bench 4	1,319	#4 of 32	Official leaderboard
BullshitBench V2	74%	#16 of 122	Official leaderboard
Humanity's Last Exam	56%	#20 of 147	Model card
Longform Creative Writing (EQ-Bench)	79.6	#17 of 119	Official leaderboard
BullshitBench	58.2%	#21 of 107	Official leaderboard
AgentIF (2026-10 setting)	64.9%	#5 of 24	Official leaderboard
SimpleQA Verified	50.6%	#27 of 84	Epoch AI
HLE-Diamond	28.3%	#14 of 18	Official leaderboard
Chat All-in-One Bench	85.2%	#1 of 1	Model card
CLIF Bench	52.4%	#1 of 1	Model card
Faithfulness (Moonshot)	85.5%	#1 of 1	Model card

Math and science

Benchmark	Score	Rank	Source
Surface Evolver Bench	95%	#2 of 26	Epoch AI
ProofBench	70%	#5 of 59	Official leaderboard
MathArena (Apex, IMO, USAMO)	65.6%	#4 of 47	Official leaderboard
GPQA Diamond	93.1%	#21 of 219	Epoch AI
HMMT 2026	97%	#4 of 31	Official leaderboard
CritPt	23.4%	#24 of 160	Epoch AI
OTIS Mock AIME 2024-2025	97.2%	#34 of 202	Epoch AI
DTBench	85.3%	#38 of 178	Official leaderboard
ARC-AGI	94.5%	#20 of 92	Official leaderboard
ProofBench v1.1	87%	#12 of 51	Official leaderboard
NatureBench	14.4%	#5 of 18	Official leaderboard
FrontierMath Tiers 1-3 v2	72.2%	#25 of 89	Epoch AI
ExplorationBench	79%	#3 of 10	Official leaderboard
SimpleBench	60.7%	#35 of 100	Official leaderboard
BixBench3	47.0	#7 of 20	Official leaderboard
AIME 2026	96.7%	#11 of 31	Official leaderboard
ARC-AGI-2	60.4%	#35 of 95	Official leaderboard
FrontierMath Tier 4 v2	39%	#27 of 71	Epoch AI
xbench-ScienceQA	71.2%	#13 of 29	Official leaderboard
PaperBenchX	7.53%	#5 of 10	Official leaderboard
Prophet Arena	85.1	#11 of 18	Official leaderboard
FrontierMath Erdős	0%	#10 of 14	Epoch AI
Terminal-Bench-Science	7.14%	#13 of 17	Official leaderboard
LatchBio Capabilities v1.0	36.4%	#21 of 27	Official leaderboard
ForecastBench	64.2	#67 of 86	Official leaderboard
FrontierPhysics	0.60%	#17 of 19	Official leaderboard

Coding

Benchmark	Score	Rank	Source
SciCode	58.7%	#4 of 47	Model card
WebDev Arena	1,674	#9 of 105	Epoch AI
WeirdML	82.6%	#10 of 112	Official leaderboard
AndroidBench	90.2%	#4 of 32	Official leaderboard
Code Arena	1,656	#16 of 124	Official leaderboard
Vibe Code Bench	85%	#17 of 115	Official leaderboard
ALE-Bench	1,524	#24 of 121	Epoch AI
SWE-Bench Pro V2	97.7%	#4 of 16	Official leaderboard
SWE Refactor Bench	19.5%	#2 of 8	Official leaderboard
DeepSWE	68.5%	#8 of 28	Official leaderboard
SWE-Marathon v1.1	48.1%	#5 of 17	Official leaderboard
Vision2Web	79%	#3 of 10	Official leaderboard
MLS-Bench	48.3%	#5 of 15	Official leaderboard
FrontierCode	44.2%	#16 of 46	Epoch AI
Image-to-WebDev Arena	1,579	#20 of 55	Official leaderboard
MirrorCode	54.4%	#11 of 28	Epoch AI
ProgramBench	77.8%	#8 of 20	Model card
Tencent WorkBuddy Bench	76.7%	#4 of 10	Official leaderboard
GBAEval	48.4%	#9 of 21	Official leaderboard
SWE-Bench Pro V2 Hard	88.2%	#7 of 16	Official leaderboard
PostTrainBench	32%	#5 of 11	Epoch AI
PostTrainBench 1.1	32%	#5 of 11	Official leaderboard
TasteVal	45.3×	#10 of 20	Official leaderboard
PostTrainBench 1.2	31.5%	#9 of 17	Official leaderboard
FrontierCS 2	20.8%	#9 of 16	Official leaderboard
FrontierSWE	25.9%	#12 of 21	Epoch AI
Vibe Code Bench 1-100	18.2%	#15 of 26	Official leaderboard
Brokk Power Ranking	7.70	#11 of 19	Official leaderboard
IOI v2 (Vals)	48.9%	#27 of 45	Official leaderboard
Vals RSI Index	0.21	#15 of 25	Official leaderboard
RSI-Exam	38.2%	#11 of 18	Official leaderboard
SWE-Marathon	42%	#7 of 11	Model card
Next.js Agent Evals	84%	#9 of 14	Official leaderboard
ViBench 2.0	0%	#6 of 9	Official leaderboard
Kimi Code Bench v2	72.9%	#2 of 3	Model card
WeirdML v3	7.54%	#17 of 25	Official leaderboard
GameDevBench	58%	#7 of 10	Official leaderboard
Code Migration	16.1%	#55 of 78	Official leaderboard
DeepRSI	820	#5 of 7	Official leaderboard
Android Bench 2.0	12%	#8 of 11	Official leaderboard
Senior SWE-Bench	20.2%	#17 of 23	Official leaderboard
RSI Bench	32.8	#6 of 8	Official leaderboard
FrontierCode 2.0	17.7%	#20 of 26	Official leaderboard
AutoResearchExam	0.43	#7 of 9	Official leaderboard
Code Migration v1.1	28.6%	#23 of 29	Official leaderboard
ScarfBench	61.3%	#8 of 10	Official leaderboard
SWE-sweep	0.60%	#8 of 10	Official leaderboard
OpenDesign Arena	65.7	#19 of 19	Official leaderboard
Coding Experience (Moonshot)	59.9%	#1 of 1	Model card
Kimi Webdev Bench	58.6%	#1 of 1	Model card

Agents and tools

Benchmark	Score	Rank	Source
DeepSearchQA	95%	#1 of 22	Model card
CEO-Bench	$22,148,357	#1 of 19	Official leaderboard
Toolathlon	76.5%	#2 of 37	Model card
MobileWorld	74.4%	#1 of 18	Official leaderboard
BrowseComp	91.2%	#6 of 77	Model card
Toolathlon-Verified	76.5%	#2 of 25	Official leaderboard
OSWorld-Verified	84.8%	#3 of 37	Model card
LisanBench	10520.7	#13 of 133	Official leaderboard
MCPMark	94.5%	#1 of 10	Model card
Terminal-Bench 2.1	88.3%	#7 of 53	Model card
GAIA	80.4%	#8 of 57	Official leaderboard
Long-Horizon-Terminal-Bench 2.0	56.8%	#3 of 15	Official leaderboard
ResearchRubrics	76.2%	#1 of 5	Model card
MCP Atlas	82.3%	#8 of 34	Official leaderboard
WitnessBench	18.8%	#3 of 12	Official leaderboard
Mystery Game Puzzles	26%	#36 of 80	Epoch AI
Chess Puzzles	37%	#38 of 83	Epoch AI
Hermes Bench	71.5%	#8 of 17	Official leaderboard
Long-Horizon-Terminal-Bench (LHTB)	37.8%	#12 of 25	Official leaderboard
τ³-bench Banking	37.1%	#12 of 25	Official leaderboard
Agent Arena	0.01	#26 of 53	Official leaderboard
ApprenticeBench	18%	#9 of 18	Official leaderboard
Swarm Bench	76.3%	#1 of 2	Model card
RecreationBench	31.4%	#6 of 10	Official leaderboard
OSWorld 2.0	58.3%	#12 of 19	Model card
AutomationBench	30.8%	#27 of 41	Model card
Time Horizon Index: KSP	10.5%	#9 of 13	Official leaderboard
tau3-bench	33.4%	#12 of 12	Model card
Agent Behavior Bench	65%	#1 of 1	Model card
Deep Research Bench (Moonshot)	90%	#1 of 1	Model card
KAET (Kimi Autonomous Execution Tasks)	83.5%	#1 of 1	Model card
SaaS-Bench	60.1%	#1 of 1	Model card

Professional work

Benchmark	Score	Rank	Source
CorpFin v2	71.6%	#3 of 119	Official leaderboard
TaxEval v2	75.7%	#12 of 129	Official leaderboard
ProfBench	57.9%	#7 of 72	Official leaderboard
LegalBench	86.2%	#16 of 138	Official leaderboard
Harvey LAB (Legal Agent Benchmark)	12.9%	#12 of 80	Official leaderboard
MedScribe	88%	#15 of 98	Official leaderboard
PolicyBench	90.5%	#9 of 50	Official leaderboard
Tax Agent Bench	33.4%	#15 of 79	Official leaderboard
Legal Research Bench	46.1%	#17 of 88	Official leaderboard
$OneMillion-Bench	35%	#3 of 13	Official leaderboard
MedCode	49.4%	#23 of 96	Official leaderboard
Public Benefits Bench	68.2%	#12 of 50	Official leaderboard
JobBench	54.3%	#4 of 16	Official leaderboard
Excel Modeling Benchmark	66.7%	#21 of 76	Official leaderboard
Immigration Bench	68.3%	#7 of 24	Official leaderboard
GDPval-AA v2	1,686	#8 of 26	Model card
Vals Legal Index	71.7%	#11 of 33	Official leaderboard
Vals Finance Index	58%	#12 of 35	Official leaderboard
Excel Modeling Benchmark v1.1	71.3%	#13 of 37	Official leaderboard
Legora BAR	102.0	#6 of 17	Official leaderboard
MortgageTax	66.3%	#32 of 86	Official leaderboard
prinzbench	47.0	#10 of 26	Official leaderboard
E-Commerce Bench	2.65	#8 of 19	Official leaderboard
Harvey LAB v1.1 (HLAB-Vals-1.1)	12.9%	#14 of 33	Official leaderboard
Ivo-micro1 ContractBench	57.1%	#3 of 7	Official leaderboard
Vals Finance Agent v2	53.1%	#35 of 80	Official leaderboard
Thrive Holdings helpdesk ticket resolution benchmark	48.1%	#7 of 15	Official leaderboard
Vending-Bench 2	$5,165	#32 of 66	Official leaderboard
TaxCalcBench v2	38%	#13 of 26	Official leaderboard
SpreadsheetBench 2	34.8%	#1 of 2	Model card
Holdings-IndividualTaxBench	72.6%	#7 of 13	Official leaderboard
FrontierFinance	46.4%	#9 of 16	Official leaderboard
APEX-Agents	47.8%	#33 of 56	Epoch AI
Vals Finance Agent v2.1	60%	#23 of 35	Official leaderboard
Agents' Last Exam	35.3%	#8 of 12	Official leaderboard
AA-Briefcase	1,548	#4 of 6	Model card
Vending-Bench Arena	$3,194	#22 of 32	Official leaderboard
Vals Index v2.2	53%	#24 of 32	Official leaderboard
OfficeQA	63.3%	#9 of 12	Model card
Build-A-Chip-Bench	0.20%	#4 of 5	Official leaderboard
Parametric CAD Bench	24.9%	#9 of 11	Official leaderboard
Epoch Automation Reports	52.4%	#5 of 6	Official leaderboard
Clipping Bench	$343	#16 of 19	Official leaderboard
BlenderWorld	46.5	#11 of 13	Official leaderboard
Quant-Arena	38.3	#8 of 9	Official leaderboard
24/7 ClawBench 2.0	48.3%	#1 of 1	Model card
DECK Bench	73.5%	#1 of 1	Model card
Finance Bench (Moonshot)	62.6%	#1 of 1	Model card
MIRA Bench	64.1%	#1 of 1	Model card
Online Experience (Moonshot)	77.9%	#1 of 1	Model card
Vals Index v1	74.7%	#1 of 1	Model card

Vision, audio and robotics

Benchmark	Score	Rank	Source
Video-MME	90%	#2 of 42	Model card
MMMU-Pro	81.6%	#7 of 72	Model card
MMVU	82.1%	#2 of 20	Model card
Math Kangaroo 2025 (MathArena)	91.3%	#3 of 22	Official leaderboard
MATH-Vision (MathVision)	94.3%	#6 of 42	Model card
BabyVision	85.7%	#4 of 27	Official leaderboard
MDPBench	83.6	#2 of 9	Official leaderboard
VoxelBench	1,916	#15 of 56	Official leaderboard
CharXiv	84.8%	#18 of 67	Model card
ArtifactArena	1,164	#8 of 21	Official leaderboard
VibeVideoBench	3.44	#5 of 9	Official leaderboard
WorldVQA	51%	#5 of 9	Model card
Code2Video Benchmark	1,481	#11 of 19	Official leaderboard
Blueprint-Bench 2	29.5%	#21 of 35	Official leaderboard
ZeroBench	23%	#15 of 25	Model card
GDP.pdf	19%	#24 of 39	Epoch AI
Furniture Assembly Benchmark	34.2%	#22 of 35	Epoch AI
DataCenterBench Tier 1	15%	#6 of 9	Official leaderboard
OmniDocBench	91.1%	#2 of 3	Model card
Humanity's Sixth Sense	25.5%	#21 of 29	Official leaderboard
ScienceVision	0.36	#8 of 11	Official leaderboard
PerceptionBench (Moonshot)	58.5%	#3 of 3	Model card
Agentic Vision Bench (Moonshot)	78.3%	#1 of 1	Model card
KWV Bench	64.7%	#1 of 1	Model card

Safety and security

Benchmark	Score	Rank	Source
Dictatorship Eval	90.3%	#3 of 20	Official leaderboard
CyberBench (Vals)	79%	#4 of 26	Official leaderboard
CyberBench v1.1 (Vals)	75.2%	#8 of 43	Official leaderboard
OncologyBench	24%	#6 of 30	Official leaderboard
Agentic BAIM-LLM Evaluation (ABLE)	78.4%	#4 of 16	Official leaderboard
Human Pathogen Capabilities Test v2 (HPCT v2)	60.2%	#27 of 71	Official leaderboard
DistressBench	73.5%	#16 of 42	Official leaderboard
KORA	52.5%	#18 of 42	Official leaderboard
DeepsecBench	17.4	#15 of 28	Official leaderboard
FORTRESS	26.6%	#36 of 66	Official leaderboard
VariantBench	31.4%	#16 of 27	Official leaderboard
PCT (Polarized Contrastive Pairs)	47.6%	#28 of 46	Official leaderboard
TxBench-Antibody-Discovery	35.3%	#18 of 28	Official leaderboard
BioSecBench-Function	32.7%	#18 of 27	Official leaderboard
scBench-Long	14.2%	#18 of 27	Official leaderboard
CheatBench	70%	#7 of 10	Official leaderboard
BioSecBench-Refusal	8.30%	#16 of 22	Official leaderboard
TxBench-Oligo-Discovery	31.3%	#21 of 28	Official leaderboard
MetagenomicsBench	27.3%	#20 of 26	Official leaderboard
BioSecBench-Surveillance	38.6%	#22 of 27	Official leaderboard
TxBench-Preclinical-Pharmacology	39.7%	#23 of 28	Official leaderboard
SpatialBench-Long	22.6%	#23 of 27	Official leaderboard
EpiBench	15.7%	#24 of 27	Official leaderboard
ExploitBench	32%	#17 of 19	Model card
Last Secure Code Benchmark	16.4%	#6 of 6	Official leaderboard
Kimi exploit development suite	38.9%	#1 of 1	Model card
UK AISI and CAISI simulated enterprise network range	17.0	#1 of 1	Model card

Other

Benchmark	Score	Rank	Source
SAGE (Vals)	52.8%	#5 of 81	Official leaderboard
Vals Multimodal Index	73.4%	#3 of 33	Official leaderboard
LMCA	52.7%	#21 of 139	Epoch AI
Conceptual Reasoning Index	62.8	#28 of 149	Official leaderboard
CAIS Text Capabilities Index	40.5	#18 of 66	Official leaderboard
SuperCLUE	70.7	#5 of 18	Official leaderboard
CAIS Vision Capabilities Index	60.6	#15 of 53	Official leaderboard
ACCoRD	70.5	#50 of 150	Official leaderboard
SAGE v1.1 (Vals)	86.8%	#11 of 28	Official leaderboard
Artificial Analysis Intelligence Index v4.1	57.1	#3 of 7	Model card
Hermes Index	39.2	#8 of 17	Official leaderboard
CAIS Risk Index	53.6	#37 of 65	Official leaderboard
LiveBench	64.6%	#18 of 28	Official leaderboard