Kimi K2.5

Kimi K2.5 ranks first on 3 of the 152 benchmarks it has scores on.

Moonshot AI Released 27 Jan 2026 Family Kimi K2 The lab's model card

155

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Moonshot AI Jan 2026

Kimi K2.5

Open weights Flagship

148.1 #71 of 271

Epoch Capabilities Index

Humanity's Last Exam 24.4% #17 /47
SWE-bench Verified 73.8% #18 /32
Terminal-Bench 2.0 43.2% #22 /43

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
MMLU-Pro	87.1%	#9 of 151	Model card
Fiction.LiveBench	86.1%	#7 of 57	Epoch AI
LMArena (Elo)	1,450	#63 of 333	Official leaderboard
Seal-0	57.4%	#1 of 5	Model card
EQ-Bench 3	1,530	#16 of 78	Official leaderboard
Longform Creative Writing (EQ-Bench)	74.9	#28 of 119	Official leaderboard
BullshitBench	47.3%	#29 of 107	Official leaderboard
BullshitBench V2	52%	#41 of 122	Official leaderboard
Humanity's Last Exam	24.4%	#17 of 47	Official leaderboard
LongBench v2	61%	#10 of 27	Model card
SimpleVQA	71.2%	#9 of 23	Model card
CL-bench	19.3%	#10 of 24	Official leaderboard
MultiChallenge	61.4%	#15 of 35	Official leaderboard
Creative Writing v3	70.4	#52 of 120	Official leaderboard
MultiNRC	35.2%	#18 of 38	Official leaderboard
Phare	67%	#30 of 62	Official leaderboard
AdvancedIF	75.6%	#3 of 6	Model card
Extended NYT Connections	69.9%	#52 of 97	Official leaderboard
CL-bench Life	13.2%	#9 of 16	Official leaderboard
WideSearch	72.7%	#10 of 17	Model card
SkateBench v2	57.2%	#13 of 21	Official leaderboard
SimpleQA Verified	34.3%	#56 of 84	Epoch AI
Document Arena	1,436	#31 of 37	Official leaderboard
Vectara Hallucination Leaderboard (2025 dataset)	14.2%	#91 of 104	Official leaderboard
LOCA-Bench	4.00%	#8 of 9	Official leaderboard

Math and science

Benchmark	Score	Rank	Source
AIME 2025	96.1%	#20 of 143	Model card
BRUMO 2025	98.3%	#7 of 42	Official leaderboard
CMIMC 2025	90.6%	#7 of 33	Official leaderboard
HMMT 2025	95.4%	#12 of 49	Model card
SMT 2025	90.6%	#10 of 40	Official leaderboard
OTIS Mock AIME 2024-2025	92.2%	#54 of 202	Epoch AI
GPQA Diamond	87.6%	#60 of 219	Epoch AI
FrontierMath	27.9%	#21 of 70	Epoch AI
AIME 2026	95.8%	#13 of 31	Official leaderboard
General365	56.4%	#10 of 23	Official leaderboard
FrontierMath Tier 4	4.20%	#26 of 55	Epoch AI
CritPt	3.14%	#76 of 160	Epoch AI
MathArena (Apex, IMO, USAMO)	8.85%	#23 of 47	Official leaderboard
IMO-AnswerBench	81.8%	#21 of 37	Model card
SimpleBench	46.8%	#57 of 100	Official leaderboard
ARC-AGI-2	11.8%	#55 of 95	Official leaderboard
ARC-AGI	65.3%	#54 of 92	Official leaderboard
HMMT 2026	87.1%	#19 of 31	Official leaderboard
HMMT Nov 2025	89.2%	#15 of 22	Official leaderboard
ResearchClawBench	14.0	#19 of 23	Official leaderboard
Echo General AI Prediction Leaderboard	920	#17 of 17	Official leaderboard

Coding

Benchmark	Score	Rank	Source
PaperBench	63.5%	#2 of 14	Model card
LiveSQLBench	32.2%	#7 of 36	Official leaderboard
StructEval	77.3%	#6 of 28	Official leaderboard
SciCode	48.7%	#11 of 47	Model card
LiveCodeBench v6	85%	#20 of 76	Model card
OJBench	57%	#3 of 8	Official leaderboard
IOI (Vals)	17.7%	#25 of 55	Official leaderboard
SWE-bench Multilingual	73%	#35 of 70	Model card
WeirdML	45.6%	#61 of 112	Official leaderboard
OpenHands Index	49.2%	#18 of 33	Official leaderboard
WebDev Arena	1,436	#58 of 105	Epoch AI
ALE-Bench	822	#67 of 121	Epoch AI
Code Arena	1,436	#69 of 124	Official leaderboard
SWE-bench Verified	73.8%	#18 of 32	Epoch AI
FrontierCode	6.86%	#8 of 12	Official leaderboard
PostTrainBench	10.3%	#17 of 23	Official leaderboard
Vibe Code Bench	17.5%	#88 of 115	Official leaderboard
TasteVal	8.60×	#16 of 20	Official leaderboard
Image-to-WebDev Arena	1,439	#46 of 55	Official leaderboard
SWE-Bench Pro	50.7%	#65 of 77	Model card
SlopCodeBench	9.69%	#12 of 14	Official leaderboard
Code Migration	6.96%	#67 of 78	Official leaderboard
SWE Atlas Refactoring	20.9%	#15 of 17	Official leaderboard
SWE Atlas	13.1%	#20 of 22	Official leaderboard
SWE Atlas Test Writing	25.8%	#21 of 22	Official leaderboard
Vals RSI Index	0.07	#25 of 25	Official leaderboard

Agents and tools

Benchmark	Score	Rank	Source
PinchBench	75.2%	#11 of 48	Official leaderboard
LisanBench	1851.7	#44 of 133	Official leaderboard
OSWorld	63.3%	#3 of 9	Epoch AI
OSWorld-Verified	63.3%	#13 of 33	Official leaderboard
GameWorld	37.4%	#5 of 12	Official leaderboard
SnakeBench	25	#85 of 183	Official leaderboard
LLM Chess	111	#61 of 124	Official leaderboard
Claw-Eval-Live	53.3%	#6 of 12	Official leaderboard
Pencil Puzzle Bench	3.30%	#27 of 53	Official leaderboard
Terminal-Bench 2.0	43.2%	#22 of 43	Official leaderboard
BrowseComp	74.9%	#40 of 77	Model card
Claw-Eval	74.9%	#14 of 23	Official leaderboard
MobileWorld	49.6%	#11 of 18	Official leaderboard
WebArena	58.9%	#5 of 8	Model card
Toolathlon	27.8%	#33 of 50	Official leaderboard
MCP Atlas	64.4%	#25 of 34	Official leaderboard
DeepSearchQA	77.1%	#17 of 22	Model card
Claw-Eval Pass^3	52.3%	#18 of 23	Official leaderboard
PinchBench 2.0	54.6%	#45 of 57	Official leaderboard
BalatroBench	5.33	#13 of 16	Official leaderboard
Toolathlon-Verified	33%	#25 of 25	Official leaderboard
ClawBench	17.6%	#6 of 6	Official leaderboard
CUA-World-Long	5.50%	#4 of 4	Official leaderboard
Swarm Bench	58.3%	#2 of 2	Model card
xbench-DeepSearch-2510	46%	#2 of 2	Official leaderboard

Professional work

Benchmark	Score	Rank	Source
CorpFin v2	68.3%	#10 of 119	Official leaderboard
Vals Finance Agent v1	49.2%	#7 of 47	Official leaderboard
TaxEval v2	74.2%	#39 of 129	Official leaderboard
ProfBench	51.8%	#24 of 72	Official leaderboard
MortgageTax	66.5%	#31 of 86	Official leaderboard
FinSearchComp	67.8%	#3 of 7	Model card
CaseLaw v2	58.7%	#28 of 53	Official leaderboard
Vals Finance Agent	50.6%	#28 of 49	Official leaderboard
prinzbench	35.0	#15 of 26	Official leaderboard
PRBench	43.8%	#26 of 44	Official leaderboard
MedCode	39.3%	#64 of 96	Official leaderboard
MedScribe	76.4%	#69 of 98	Official leaderboard
SpreadsheetBench 2	14.6%	#6 of 8	Official leaderboard
Legal Research Bench	15.9%	#67 of 88	Official leaderboard
Vending-Bench 2	$1,198	#51 of 66	Official leaderboard
Harvey LAB (Legal Agent Benchmark)	0%	#65 of 80	Official leaderboard
Vals Finance Agent v2	35.8%	#69 of 80	Official leaderboard
Excel Modeling Benchmark	28.5%	#66 of 76	Official leaderboard
FORTE	55.6%	#19 of 21	Official leaderboard

Vision, audio and robotics

Benchmark	Score	Rank	Source
InfographicVQA	92.6%	#1 of 22	Model card
MathVista	90.1%	#3 of 64	Model card
MMMU	84.3%	#5 of 97	Model card
BLINK	78.9%	#3 of 28	Model card
OCRBench	92.3%	#3 of 28	Model card
VideoMME v2 (with subtitles)	40.7%	#4 of 31	Official leaderboard
Video-MMMU	86.6%	#6 of 30	Model card
MMVU	80.4%	#4 of 20	Model card
MMVP	87%	#1 of 5	Model card
MMMU-Pro	78.5%	#18 of 72	Model card
LongVideoBench	79.8%	#3 of 11	Model card
Video-MME	87.4%	#12 of 42	Model card
Vision Arena	1,252	#42 of 136	Official leaderboard
MDPBench	77.5	#3 of 9	Official leaderboard
MATH-Vision (MathVision)	84.2%	#15 of 42	Model card
OmniDocBench v1.5	89.3	#4 of 10	Official leaderboard
LVBench	75.9%	#12 of 25	Model card
Math Kangaroo 2025 (MathArena)	80.6%	#11 of 22	Official leaderboard
Claw-Eval-MM	50.2%	#6 of 11	Official leaderboard
MotionBench	70.4%	#5 of 9	Model card
VISTA	41.9%	#31 of 54	Official leaderboard
BabyVision	36.5%	#16 of 27	Official leaderboard
CharXiv	77.5%	#40 of 67	Model card
EnigmaEval	3.38%	#26 of 41	Epoch AI
ZeroBench	9.00%	#21 of 25	Model card
OmniDocBench v1.6	84.5	#6 of 7	Official leaderboard
WorldVQA	46.3%	#9 of 9	Model card

Safety and security

Benchmark	Score	Rank	Source
Agentic BAIM-LLM Evaluation (ABLE)	82.5%	#3 of 16	Official leaderboard
BinaryAudit	24.6%	#10 of 20	Official leaderboard
KORA	38.5%	#23 of 42	Official leaderboard
SMDD-Bench	30.3%	#4 of 7	Official leaderboard
CyberGym	41.3%	#20 of 33	Official leaderboard
FORTRESS	41.1%	#45 of 66	Official leaderboard
Human Pathogen Capabilities Test v2 (HPCT v2)	43.4%	#53 of 71	Official leaderboard
SEC-bench Pro	2.20%	#5 of 6	Official leaderboard
SEC-bench Pro 260617	2.30%	#5 of 6	Official leaderboard

Other

Benchmark	Score	Rank	Source
SAGE (Vals)	49.9%	#19 of 81	Official leaderboard
CAIS Text Capabilities Index	24.9	#43 of 66	Official leaderboard
CAIS Risk Index	65.5	#57 of 65	Official leaderboard

Moonshot AI's models over time

Moonshot AI's models in color, against every other model on the Epoch Capabilities Index, by release date.

Its launch post

The benchmarks the lab chose to headline, with the scores it gave.

Kimi K2.5 model card (Hugging Face) 27 Jan 2026

Best of HLE-Full / HLE-Full (w/ tools) 50.2
AIME 2025 96.1
HMMT 2025 (Feb) 95.4
IMO-AnswerBench 81.8
GPQA-Diamond 87.6
MMLU-Pro 87.1
MMMU-Pro 78.5
CharXiv (RQ) 77.5
MathVision 84.2
MathVista (mini) 90.1
ZeroBench 9
OCRBench 92.3
OmniDocBench 1.5 88.8
InfoVQA (val) 92.6
SimpleVQA 71.2
WorldVQA 46.3
VideoMMMU 86.6
MMVU 80.4
MotionBench 70.4
VideoMME 87.4
LongVideoBench 79.8
LVBench 75.9
SWE-Bench Verified 76.8
SWE-Bench Pro 50.7
SWE-Bench Multilingual 73
Terminal Bench 2.0 50.8
PaperBench 63.5
CyberGym 41.3
SciCode 48.7
OJBench (cpp) 57.4
LiveCodeBench (v6) 85
Longbench v2 61
BrowseComp (w/ctx manage) 74.9
WideSearch (item-f1) 72.7
DeepSearchQA 77.1
FinSearchCompT2&T3 67.8
Seal-0 57.4

Its system card

Every result in the longer documents the lab published with it. 11 of them aren't in the launch post.

Kimi K2.5: Visual Agentic Intelligence (arXiv 2602.02276) 30 Jan 2026

Capabilities