Head-to-head comparison

Kimi K3 vs. GLM-5.2

Compare published benchmark results from matching versions and cohorts, API pricing, and technical specifications. Any documented test setup differences remain visible.

shared result series: 179
distinct benchmarks: 20
latest retrieval: 2026-09-04

First modelSecond model

Specifications

General specifications

Release date, availability, context window, architecture, and published parameter counts at a glance. Undisclosed details remain marked as unavailable.

Specifications for Kimi K3 and GLM-5.2
Attribute	Kimi K3	GLM-5.2
Provider	Moonshot AI	Z.ai
Released	Jul 16, 2026	Jun 16, 2026
Availability	Active	Open Source
Model type	Open Weights	Open Weights
Parameters	2.8T, 104B active	744B, 40B active
Architecture	Mixture of Experts	Mixture of Experts
Context window	1,048,576 tokens	1,000,000 tokens
Knowledge cutoff	Not published	Not published
Technical sources	Model	Model, Context

Pricing

API pricing compared

The table compares official standard API rates per 1M tokens, including published cache pricing. Batch, fast, priority, regional surcharges, tool calls, and cloud platform pricing are excluded.

API pricing for Kimi K3 and GLM-5.2
Price type	Kimi K3	GLM-5.2
API input	$3	$1.4
API output	$15	$4.4
Cache read	$0.3	$0.26
Cache write (5 min.)	Not listed	Not listed
Cache write (1 hr.)	Not listed	Not listed
Price verified	07/30/2026 Source	07/30/2026 Source

Kimi K3GLM-5.2

API inputUSD per 1M tokens, base tier

Kimi K3$3

GLM-5.2$1.4

API outputUSD per 1M tokens, base tier

Kimi K3$15

GLM-5.2$4.4

Cache readUSD per 1M tokens, base tier

Kimi K3$0.3

GLM-5.2$0.26

Key to the marks:better valuebetter, but not conclusivetie

Pricing by provider

Direct developer pricing and provider endpoint tariffs routed through OpenRouter are listed separately. Prices are in USD per 1M tokens. Context-dependent direct tiers plus regional, Flex, Priority, and other OpenRouter tariffs remain individually identifiable.

Provider pricing for Kimi K3 and GLM-5.2
Model	Provider	Purchase route and tariff	Input per 1M tokens	Output per 1M tokens	Source
Kimi K3	Moonshot AI	Direct from developerStandard API	$3	$15	Source07/30/2026
	Alibaba	Via OpenRouteralibaba	$3.45	$17.25	Source09/07/2026
	BaseTen	Via OpenRouterbaseten/fp8	$3	$15	Source09/07/2026
	Chutes	Via OpenRouterchutes/mxfp4	$3	$15	Source09/07/2026
	DeepInfra	Via OpenRouterdeepinfra/bf16	$2.85	$14.25	Source09/07/2026
	DigitalOcean	Via OpenRouterdigitalocean	$2.85	$14.25	Source09/07/2026
	Fireworks	Via OpenRouterfireworks	$3	$15	Source09/07/2026
	Fireworks	Via OpenRouterfireworks/fast	$4.5	$22.5	Source09/07/2026
	Fireworks	Via OpenRouterfireworks/us	$3.3	$16.5	Source09/07/2026
	Makora	Via OpenRoutermakora	Lowest listed price: $2.55	Lowest listed price: $12.75	Source09/07/2026
	Modal	Via OpenRoutermodal/mxfp4	$3	$15	Source09/07/2026
	Moonshot AI	Via OpenRoutermoonshotai/mxfp4	$3	$15	Source09/07/2026
	Morph	Via OpenRoutermorph/fast	$6	$22.5	Source09/07/2026
	Morph	Via OpenRoutermorph/fp4	$2.6	$14	Source09/07/2026
	Parasail	Via OpenRouterparasail/fp4	$3	$15	Source09/07/2026
	Phala	Via OpenRouterphala	$3	$15	Source09/07/2026
	Sail Research	Via OpenRoutersail-research/fp4	$2.6	$13	Source09/07/2026
	Together	Via OpenRoutertogether	$3	$15	Source09/07/2026
	Wafer	Via OpenRouterwafer	$3	Lowest listed price: $12.75	Source09/07/2026
GLM-5.2	Z.ai	Direct from developerStandard API	$1.4	$4.4	Source07/30/2026
	Alibaba	Via OpenRouteralibaba/fast	$2.31	$7.26	Source09/07/2026
	Alibaba	Via OpenRouteralibaba/fp8	$0.966	$3.036	Source09/07/2026
	Ambient	Via OpenRouterambient/fp8	$0.6	$2	Source09/07/2026
	AtlasCloud	Via OpenRouteratlas-cloud/fp8	$0.938	$2.948	Source09/07/2026
	Baidu	Via OpenRouterbaidu/fp8	$1.4	$4.4	Source09/07/2026
	BaseTen	Via OpenRouterbaseten/fast#1	$2.1	$6.6	Source09/07/2026
	BaseTen	Via OpenRouterbaseten/fast#2	$2.1	$6.6	Source09/07/2026
	BaseTen	Via OpenRouterbaseten/fp8#1	$1.4	$4.4	Source09/07/2026
	BaseTen	Via OpenRouterbaseten/fp8#2	$1.4	$4.4	Source09/07/2026
	Cloudflare	Via OpenRoutercloudflare	$1.4	$4.4	Source09/07/2026
	CoreWeave	Via OpenRoutercoreweave/fp4	$0.76	$2.42	Source09/07/2026
	Crusoe	Via OpenRoutercrusoe/fp8	$1.4	$4.4	Source09/07/2026
	Decart	Via OpenRouterdecart/fast	$1.26	$3.96	Source09/07/2026
	Decart	Via OpenRouterdecart/fp4	$0.684	$2.28	Source09/07/2026
	DeepInfra	Via OpenRouterdeepinfra/fp4	Lowest listed price: $0.4875	Lowest listed price: $1.56	Source09/07/2026
	DigitalOcean	Via OpenRouterdigitalocean	$0.7	$2.2	Source09/07/2026
	Fireworks	Via OpenRouterfireworks	$1.4	$4.4	Source09/07/2026
	Fireworks	Via OpenRouterfireworks/fast	$2.1	$6.6	Source09/07/2026
	Fireworks	Via OpenRouterfireworks/fast-us	$2.1	$6.6	Source09/07/2026
	Friendli	Via OpenRouterfriendli	$1.4	$4.4	Source09/07/2026
	GMICloud	Via OpenRoutergmicloud/fp8	$1.05	$3.3	Source09/07/2026
	Inceptron	Via OpenRouterinceptron/fp4	$1.25	$2.99	Source09/07/2026
	Mistral	Via OpenRoutermistral	$1.4	$4.4	Source09/07/2026
	Mistral	Via OpenRoutermistral/eu	$1.54	$4.84	Source09/07/2026
	Mistral	Via OpenRoutermistral/zdr	$1.4	$4.4	Source09/07/2026
	Novita	Via OpenRouternovita/fp8	$0.5586	$1.7556	Source09/07/2026
	Parasail	Via OpenRouterparasail/fp4	$1.4	$4.4	Source09/07/2026
	Phala	Via OpenRouterphala/fp8	$1.26	$3	Source09/07/2026
	SiliconFlow	Via OpenRoutersiliconflow/fp8	$1.19	$3.74	Source09/07/2026
	StreamLake	Via OpenRouterstreamlake/fp8	$0.5446	$1.7116	Source09/07/2026
	Together	Via OpenRoutertogether	$1.4	$4.4	Source09/07/2026
	Venice	Via OpenRoutervenice/fp8	$1.4	$4.4	Source09/07/2026
	Z.AI	Via OpenRouterz-ai/fp8	$1.4	$4.4	Source09/07/2026

Performance

Shared benchmarks

Only published results with the same benchmark version, task, metric, and comparison cohort are paired. Different reasoning levels, agents, harnesses, or output limits appear directly in the table.

Capability profile from matched benchmarks

Each axis averages directly matched benchmark families. A value of 100 means the stronger value within this pair, not a universal quality score.

Kimi K3

GLM-5.2

Coding: 90.4 / 77.8 (6 benchmark families)

Reasoning: 100 / 95.1 (2 benchmark families)

Mathematics: 100 / 98.1 (1 benchmark family)

Knowledge: 100 / 98.6 (2 benchmark families)

Finance: 100 / 94.3 (5 benchmark families)

Legal: 100 / 83 (4 benchmark families)

Agentic tasks: 100 / 83.8 (1 benchmark family)

Cybersecurity: 100 / 97.9 (1 benchmark family)

49 matched rows from 20 benchmark families were considered. The radar shows 8 of 11 comparable categories.

Kimi K3GLM-5.2

Code MigrationOverall. Higher is better. No documented setup difference

Kimi K316.099 %

GLM-5.237.87 %

CorpFin v2Overall. Higher is better. No documented setup difference

Kimi K371.562 %

GLM-5.266.123 %

CyberBenchOverall. Higher is better. No documented setup difference

Kimi K379.026 %

GLM-5.277.359 %

EMBOverall. Higher is better. No documented setup difference

Kimi K366.399 %

GLM-5.261.532 %

Finance Agent (v2)Overall. Higher is better. No documented setup difference

Kimi K354.36 %

GLM-5.249.699 %

GPQA DiamondOverall. Higher is better. No documented setup difference

Kimi K392.929 %

GLM-5.285.606 %

Harvey's Legal Agent BenchmarkOverall · Task fully resolved. Higher is better. No documented setup difference

Kimi K310.833 %

GLM-5.27.083 %

Legal Research BenchOverall · All-pass. Higher is better. No documented setup difference

Kimi K344.231 %

GLM-5.231.25 %

LegalBenchOverall. Higher is better. No documented setup difference

Kimi K386.016 %

GLM-5.284.073 %

LiveCodeBench v6Overall. Higher is better. No documented setup difference

Kimi K387.185 %

GLM-5.269.5 %

LMArena Text, Style ControlOverall. Higher is better. No documented setup difference

Kimi K31,489.017

GLM-5.21,471.886

MedCodeOverall. Higher is better. No documented setup difference

Kimi K348.884 %

GLM-5.240.771 %

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for Kimi K3 and GLM-5.2
Benchmark and task	Kimi K3	GLM-5.2	Source
Code MigrationOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	16.099 %	Winner37.87 %	Vals AI Source details Data as of: 09/03/2026
CorpFin v2Overall Test details No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better	Winner71.562 %	66.123 %	Vals AI Source details Data as of: 08/12/2026
CyberBenchOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner79.026 %	77.359 %	Vals AI Source details Data as of: 09/01/2026
EMBOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner66.399 %	61.532 %	Vals AI Source details Data as of: 09/03/2026
Finance Agent (v2)Overall Test details No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better	Winner54.36 %	49.699 %	Vals AI Source details Data as of: 09/03/2026
GPQA DiamondOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner92.929 %	85.606 %	Vals AI Source details Data as of: 09/01/2026
Harvey's Legal Agent BenchmarkOverall · Task fully resolved Test details No documented difference in the test setupVersion: 1Metric: task resolution rateScoring: Higher is better	Winner10.833 %	7.083 %	Vals AI Source details Data as of: 09/03/2026
Legal Research BenchOverall · All-pass Test details No documented difference in the test setupVersion: 1Metric: all-pass rateScoring: Higher is better	Winner44.231 %	31.25 %	Vals AI Source details Data as of: 09/03/2026
LegalBenchOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner86.016 %	84.073 %	Vals AI Source details Data as of: 09/01/2026
LiveCodeBench v6Overall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner87.185 %	69.5 %	Vals AI Source details Data as of: 09/01/2026
LMArena Text, Style ControlOverall Test details No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterKimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style control GLM-5.2: Snapshot: glm-5.2-max, Reasoning: max, Harness: Arena text, style controlKimi K3: Style-controlled Arena rating from 17,895 anonymous pairwise votes.GLM-5.2: Style-controlled Arena rating from 33,758 anonymous pairwise votes.	Winner1,489.017 95% confidence interval: ±5.47The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	1,471.886 95% confidence interval: ±4.729The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	LMArena Source details Data as of: 09/01/2026
MedCodeOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner48.884 %	40.771 %	Vals AI Source details Data as of: 09/03/2026
MedScribeOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner87.958 %	83.534 %	Vals AI Source details Data as of: 09/03/2026
MMLU-ProOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner87.973 %	86.714 %	Vals AI Source details Data as of: 09/01/2026
ProgramBenchOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner2 %	0.5 %	Vals AI Source details Data as of: 09/01/2026
SWE-bench VerifiedOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner93.4 %	82.8 %	Vals AI Source details Data as of: 09/01/2026
TaxEval v2Overall Test details No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better	Winner75.716 %	73.344 %	Vals AI Source details Data as of: 09/01/2026
Terminal-Bench 2.1Overall Test details No documented difference in the test setupVersion: 2.1Metric: accuracyScoring: Higher is better	Winner80.899 %	67.79 %	Vals AI Source details Data as of: 09/03/2026
Vals IndexOverall Test details No documented difference in the test setupVersion: 2Metric: weighted index scoreScoring: Higher is better	Winner57.813 %	53.122 %	Vals AI Source details Data as of: 09/03/2026
Vibe Code Bench v1.1Overall Test details No documented difference in the test setupVersion: 1.1Metric: accuracyScoring: Higher is betterSettings: Harness: OpenHands	Winner84.963 %	63.961 %	Vals AI Source details Data as of: 09/03/2026
CorpFin v2Exact Pages Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value69.347 %	63.636 %	Vals AI Source details Data as of: 08/06/2026
CyberBenchPatch Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value83.051 %	81.034 %	Vals AI Source details Data as of: 08/03/2026
EMBComps Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value72.685 %	54.922 %	Vals AI Source details Data as of: 08/06/2026
Finance Agent (v2)Adjustments Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value51.461 %	46.305 %	Vals AI Source details Data as of: 08/06/2026
GPQA DiamondFew-Shot CoT Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value93.434 %	83.333 %	Vals AI Source details Data as of: 08/09/2026
Harvey's Legal Agent BenchmarkAntitrust Competition · Criteria passed Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: criteria pass rateScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	No direct winner87.319 %	No direct winner87.319 %	Vals AI Source details Data as of: 08/07/2026
Legal Research BenchAdministrative / Regulatory · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value54.545 %	39.394 %	Vals AI Source details Data as of: 08/06/2026
LegalBenchConclusion Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value90.733 %	88.587 %	Vals AI Source details Data as of: 08/09/2026
LiveCodeBench v6Easy Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value98.758 %	97.205 %	Vals AI Source details Data as of: 08/09/2026
LMArena Text, Style ControlBusiness, management and finance Test details No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterKimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style control GLM-5.2: Snapshot: glm-5.2-max, Reasoning: max, Harness: Arena text, style controlKimi K3: Style-controlled Arena rating from 3,520 anonymous pairwise votes.GLM-5.2: Style-controlled Arena rating from 6,609 anonymous pairwise votes.	Winner1,489.216 95% confidence interval: ±10.41The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	1,460.145 95% confidence interval: ±8.046The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	LMArena Source details Data as of: 09/01/2026
MMLU-ProBiology Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	93.863 %	Higher measured value94.003 %	Vals AI Source details Data as of: 08/09/2026
ProgramBenchAlmost Resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 1, Output limit: 64,000 tokens	Higher measured value17.5 %	9.5 %	Vals AI Source details Data as of: 08/09/2026
SWE-bench Verified<15 min fix Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value92.268 %	88.144 %	Vals AI Source details Data as of: 08/08/2026
TaxEval v2Correctness Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value65.331 %	62.061 %	Vals AI Source details Data as of: 08/09/2026
Terminal-Bench 2.1Easy Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 1, Output limit: 48,000 tokens	91.667 %	Higher measured value100 %	Vals AI Source details Data as of: 08/06/2026
Vals IndexCorpFin v2 Test details Test setup not documented as identicalNo direct winnerVersion: 1.2Metric: benchmark scoreScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value72.611 %	66.317 %	Vals AI Source details Data as of: 08/09/2026
CorpFin v2Max Fitting Context Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value72.727 %	68.415 %	Vals AI Source details Data as of: 08/06/2026
CyberBenchPoC Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value75 %	73.684 %	Vals AI Source details Data as of: 08/03/2026
EMBDataroom Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value85.13 %	83.444 %	Vals AI Source details Data as of: 08/06/2026
Finance Agent (v2)All-Pass Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value41.69 %	38.006 %	Vals AI Source details Data as of: 08/06/2026
GPQA DiamondZero-Shot CoT Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value92.424 %	87.879 %	Vals AI Source details Data as of: 08/09/2026
Harvey's Legal Agent BenchmarkAntitrust Competition · Task fully resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: task resolution rateScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	No direct winner0 %	No direct winner0 %	Vals AI Source details Data as of: 08/07/2026
Legal Research BenchAdministrative / Regulatory · Weighted partial credit Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: weighted partial-credit scoreScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value90.731 %	85.95 %	Vals AI Source details Data as of: 08/06/2026
LegalBenchInterpretation Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value82.612 %	82.17 %	Vals AI Source details Data as of: 08/09/2026
LiveCodeBench v6Hard Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value74.286 %	37.143 %	Vals AI Source details Data as of: 08/09/2026
LMArena Text, Style ControlChinese Test details No documented difference in the test setupNo clear leadVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterKimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style control GLM-5.2: Snapshot: glm-5.2-max, Reasoning: max, Harness: Arena text, style controlKimi K3: Style-controlled Arena rating from 1,168 anonymous pairwise votes.GLM-5.2: Style-controlled Arena rating from 2,084 anonymous pairwise votes.	No clear lead1,528.536 95% confidence interval: ±17.657The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	1,516.828 95% confidence interval: ±13.481The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	LMArena Source details Data as of: 09/01/2026
MMLU-ProBusiness Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value92.015 %	91.888 %	Vals AI Source details Data as of: 08/09/2026
ProgramBenchFully Resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 1, Output limit: 64,000 tokens	Higher measured value2 %	0.5 %	Vals AI Source details Data as of: 08/09/2026
SWE-bench Verified>4 hours Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	No direct winner66.667 %	No direct winner66.667 %	Vals AI Source details Data as of: 08/08/2026
TaxEval v2Stepwise Reasoning Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value86.1 %	84.628 %	Vals AI Source details Data as of: 08/09/2026
Terminal-Bench 2.1Hard Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 1, Output limit: 48,000 tokens	Higher measured value74.444 %	51.111 %	Vals AI Source details Data as of: 08/06/2026
Vals IndexFinance Agent v2 Test details Test setup not documented as identicalNo direct winnerVersion: 1.2Metric: benchmark scoreScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value55.882 %	50.83 %	Vals AI Source details Data as of: 08/09/2026
CorpFin v2Shared Max Context Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value72.611 %	66.317 %	Vals AI Source details Data as of: 08/06/2026
EMBDCF Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value58.877 %	52.968 %	Vals AI Source details Data as of: 08/06/2026
Finance Agent (v2)Comparables Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value46.691 %	44.78 %	Vals AI Source details Data as of: 08/06/2026
Harvey's Legal Agent BenchmarkArbitration International Dispute Resolution · Criteria passed Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: criteria pass rateScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value92.115 %	91.756 %	Vals AI Source details Data as of: 08/07/2026
Legal Research BenchBusiness & Commercial · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value46.512 %	27.907 %	Vals AI Source details Data as of: 08/06/2026
LegalBenchIssue Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value91.099 %	89.506 %	Vals AI Source details Data as of: 08/09/2026
LiveCodeBench v6Medium Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value88.512 %	74.151 %	Vals AI Source details Data as of: 08/09/2026
LMArena Text, Style ControlCoding Test details No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterKimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style control GLM-5.2: Snapshot: glm-5.2-max, Reasoning: max, Harness: Arena text, style controlKimi K3: Style-controlled Arena rating from 4,669 anonymous pairwise votes.GLM-5.2: Style-controlled Arena rating from 9,517 anonymous pairwise votes.	Winner1,541.739 95% confidence interval: ±9.193The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	1,509.051 95% confidence interval: ±7.106The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	LMArena Source details Data as of: 09/01/2026
MMLU-ProChemistry Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value91.784 %	91.608 %	Vals AI Source details Data as of: 08/09/2026
ProgramBenchRaw Pass Rate Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 1, Output limit: 64,000 tokens	Higher measured value62.768 %	62.605 %	Vals AI Source details Data as of: 08/09/2026
SWE-bench Verified1-4 hours Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 131,072 tokens	Higher measured value92.857 %	66.667 %	Vals AI Source details Data as of: 08/08/2026
Terminal-Bench 2.1Medium Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens GLM-5.2: Reasoning: max, Temperature: 1, Top P: 1, Output limit: 48,000 tokens	Higher measured value83.636 %	74.545 %	Vals AI Source details Data as of: 08/06/2026
Key to the marks:better valuebetter, but not conclusivetie

How to read this comparison

Winning one benchmark is not an overall verdict

Your workload, budget, and required context length matter most. A coding benchmark says little about visual understanding or agent performance.

The standard price table uses direct developer rates. The provider comparison labels OpenRouter endpoints, regions, and special tariffs separately.

Every result links to its measurement source and retrieval date. Labels distinguish vendor reports, official benchmark leaderboards, and independent evaluations. Disputed or archived results are excluded.

More matchups

Compare Kimi K3 and GLM-5.2 with other leading models using the same data and benchmark logic.

View the complete LLM comparison