Head-to-head comparison

GPT-6 Astra vs. Kimi K3

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

shared result series: 206
distinct benchmark metrics: 24
latest retrieval: 2026-10-15

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 GPT-6 Astra and Kimi K3
Attribute	GPT-6 Astra	Kimi K3
Provider	OpenAI	Moonshot AI
Released	Sep 3, 2026	Jul 16, 2026
Availability	Limited access	Active
Model type	Proprietary	Open Weights
Parameters	Not published	2.8T, 104B active
Architecture	Not published	Mixture of Experts
Context window	1,050,000 tokens	1,048,576 tokens
Knowledge cutoff	April 30, 2026	Not published
Technical sources	Model, Context	Model

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 GPT-6 Astra and Kimi K3
Price type	GPT-6 Astra	Kimi K3
API input	$10 ≤272K / $20 >272K	$3
API output	$50 ≤272K / $75 >272K	$15
Cache read	$1 ≤272K / $2 >272K	$0.3
Cache write (5 min.)	$12.5 ≤272K / $25 >272K	Not listed
Cache write (1 hr.)	Not listed	Not listed
Price verified	09/22/2026 Source	09/22/2026 Source

GPT-6 AstraKimi K3

API inputUSD per 1M tokens, base tier

GPT-6 Astra$10

Kimi K3$3

API outputUSD per 1M tokens, base tier

GPT-6 Astra$50

Kimi K3$15

Cache readUSD per 1M tokens, base tier

GPT-6 Astra$1

Kimi K3$0.3

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 GPT-6 Astra and Kimi K3
Model	Provider	Purchase route and tariff	Input per 1M tokens	Output per 1M tokens	Source
GPT-6 Astra	OpenAI	Direct from developerUp to 272,000 context tokens	$10	$50	Source09/22/2026
	OpenAI	Direct from developerAbove 272,000 context tokens	$20	$75	Source09/22/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/us-west-2	$11	$55	Source10/19/2026
	Azure	Via OpenRouterazure	$10	$50	Source10/19/2026
	Azure	Via OpenRouterazure/us	$11	$55	Source10/19/2026
	OpenAI	Via OpenRouteropenai	$10	$50	Source10/19/2026
	OpenAI	Via OpenRouteropenai/fast	$20	$100	Source10/19/2026
	OpenAI	Via OpenRouteropenai/flex	Lowest listed price: $5	Lowest listed price: $25	Source10/19/2026
	OpenAI	Via OpenRouteropenai/ultrafast	$60	$300	Source10/19/2026
Kimi K3	Moonshot AI	Direct from developerStandard API	$3	$15	Source09/22/2026
	AkashML	Via OpenRouterakashml/fp4	$1.2	$14	Source10/19/2026
	Alibaba	Via OpenRouteralibaba	$3.45	$17.25	Source10/19/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/us-east-2	$3	$15	Source10/19/2026
	BaseTen	Via OpenRouterbaseten/fp8	$3	$15	Source10/19/2026
	Chutes	Via OpenRouterchutes/mxfp4	$1.85	$9.5	Source10/19/2026
	Decart	Via OpenRouterdecart/mxfp4	$1.65	Lowest listed price: $8.25	Source10/19/2026
	DeepInfra	Via OpenRouterdeepinfra/mxfp4	$2.85	$14.25	Source10/19/2026
	DigitalOcean	Via OpenRouterdigitalocean	$2.55	$12.95	Source10/19/2026
	Fireworks	Via OpenRouterfireworks	$3	$15	Source10/19/2026
	Fireworks	Via OpenRouterfireworks/fast	$4.5	$22.5	Source10/19/2026
	Fireworks	Via OpenRouterfireworks/us	$4.5	$22.5	Source10/19/2026
	InferenceNet	Via OpenRouterinference-net/fast	$2.49	$15	Source10/19/2026
	InferenceNet	Via OpenRouterinference-net/fp4	Lowest listed price: $0.249	$11.936	Source10/19/2026
	Makora	Via OpenRoutermakora/fp4	$1.53	$12.75	Source10/19/2026
	Modal	Via OpenRoutermodal/mxfp4	$3	$15	Source10/19/2026
	Moonshot AI	Via OpenRoutermoonshotai/mxfp4	$3	$15	Source10/19/2026
	Morph	Via OpenRoutermorph/fp8	$0.57	$14.9	Source10/19/2026
	Parasail	Via OpenRouterparasail/fast	$3	$15	Source10/19/2026
	Phala	Via OpenRouterphala	$2.55	$12.75	Source10/19/2026
	Relace	Via OpenRouterrelace/fp4	$0.25	$14	Source10/19/2026
	Sail Research	Via OpenRoutersail-research/fp4	$1.2	$9	Source10/19/2026
	Sail Research	Via OpenRoutersail-research/fp4/us	$1.2	$9	Source10/19/2026
	Together	Via OpenRoutertogether	$3	$15	Source10/19/2026
	Wafer	Via OpenRouterwafer	$0.28	$16	Source10/19/2026
	Wafer	Via OpenRouterwafer/us	$0.36	$11.9	Source10/19/2026

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.

GPT-6 Astra

Kimi K3

Coding: 100 / 96.1 (2 benchmark families)

Reasoning: 98.9 / 100 (1 benchmark family)

Mathematics: 98.9 / 100 (1 benchmark family)

Knowledge: 99 / 100 (1 benchmark family)

Finance: 99 / 100 (1 benchmark family)

Legal: 98.4 / 100 (1 benchmark family)

Health: 97.1 / 100 (1 benchmark family)

General: 98.7 / 100 (1 benchmark family)

34 matched rows from 2 benchmark families were considered. The radar shows 8 of 8 comparable categories.

GPT-6 AstraKimi K3

BioMysteryBenchOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra79.259 %

Kimi K372.593 %

Code MigrationOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra67.74 %

Kimi K316.099 %

CyberBench v1.1Overall. Higher is better. Test setup not documented as identical

GPT-6 Astra41.072 %

Kimi K375.238 %

EMBOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra71.7 %

Kimi K366.677 %

Finance Agent (v2)Overall. Higher is better. Test setup not documented as identical

GPT-6 Astra53.54 %

Kimi K353.114 %

Harvey's Legal Agent BenchmarkOverall · Task fully resolved. Higher is better. Test setup not documented as identical

GPT-6 Astra5.417 %

Kimi K312.917 %

IOIOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra100 %

Kimi K348.944 %

Legal Research BenchOverall · All-pass. Higher is better. Test setup not documented as identical

GPT-6 Astra39.423 %

Kimi K346.154 %

LMArena Code Arena WebDevOverall. Higher is better. No documented setup difference

GPT-6 Astra1,780.244

Kimi K31,655.295

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

GPT-6 Astra1,476.716

Kimi K31,488.095

MedCodeOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra48.486 %

Kimi K349.357 %

MedScribeOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra87.908 %

Kimi K388.048 %

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for GPT-6 Astra and Kimi K3
Benchmark and task	GPT-6 Astra	Kimi K3	Source
BioMysteryBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: Terminus 2, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Harness: Terminus 2, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value79.259 %	72.593 %	Vals AI Source details Data as of: 10/12/2026
Code MigrationOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value67.74 %	16.099 %	Vals AI Source details Data as of: 10/12/2026
CyberBench v1.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Output limit: 262,144 tokens	41.072 %	Higher measured value75.238 %	Vals AI Source details Data as of: 10/12/2026
EMBOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value71.7 %	66.677 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value53.54 %	53.114 %	Vals AI Source details Data as of: 10/12/2026
Harvey's Legal Agent BenchmarkOverall · Task fully resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: task resolution rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	5.417 %	Higher measured value12.917 %	Vals AI Source details Data as of: 10/12/2026
IOIOverall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value100 %	48.944 %	Vals AI Source details Data as of: 10/14/2026
Legal Research BenchOverall · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	39.423 %	Higher measured value46.154 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevOverall Test details No documented difference in the test setupVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDev Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevGPT-6 Astra: Arena rating from 5,746 published pairwise votes.Kimi K3: Arena rating from 18,644 published pairwise votes.	Winner1,780.244 95% confidence interval: ±10.541The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,655.295 95% confidence interval: ±6.069The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Text, Style ControlOverall Test details No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 11,674 published pairwise votes.Kimi K3: Arena rating from 32,036 published pairwise votes.	1,476.716 95% confidence interval: ±6.385The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	Winner1,488.095 95% confidence interval: ±4.609The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
MedCodeOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	48.486 %	Higher measured value49.357 %	Vals AI Source details Data as of: 10/13/2026
MedScribeOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	87.908 %	Higher measured value88.048 %	Vals AI Source details Data as of: 10/13/2026
ProgramBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: fully resolved rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value5.5 %	2 %	Vals AI Source details Data as of: 10/12/2026
ProofBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Output limit: 262,144 tokens	Higher measured value99 %	87 %	Vals AI Source details Data as of: 10/12/2026
SAGEOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	46.367 %	Higher measured value52.783 %	Vals AI Source details Data as of: 10/08/2026
SWE-bench MultimodalOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value43.333 %	35.833 %	Vals AI Source details Data as of: 10/13/2026
Tax Agent BenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	20.717 %	Higher measured value33.368 %	Vals AI Source details Data as of: 10/12/2026
Terminal-Bench 2.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value87.266 %	80.899 %	Vals AI Source details Data as of: 09/28/2026
Terminal-Bench 4.0Overall Test details Test setup not documented as identicalNo direct winnerVersion: 4.0Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value59.596 %	17.172 %	Vals AI Source details Data as of: 10/13/2026
Terminal-Bench ScienceOverall Test details Test setup not documented as identicalNo direct winnerVersion: 0.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: Mini-SWE-agent, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Harness: Mini-SWE-agent, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value62.857 %	1.429 %	Vals AI Source details Data as of: 10/14/2026
Time Horizon Index: KSPOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: progressScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value90.5 %	10.5 %	Vals AI Source details Data as of: 10/07/2026
Vals IndexOverall Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: weighted index scoreScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value63.125 %	50.297 %	Vals AI Source details Data as of: 10/13/2026
Vibe Code Bench 1-100Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.0Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Harness: OpenHands, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value27.644 %	18.244 %	Vals AI Source details Data as of: 10/14/2026
Vibe Code Bench v1.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Harness: OpenHands, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value89.594 %	84.968 %	Vals AI Source details Data as of: 10/12/2026
BioMysteryBenchHard Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: Terminus 2, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Harness: Terminus 2, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	37.255 %	Higher measured value41.176 %	Vals AI Source details Data as of: 10/12/2026
Code MigrationCLI Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value67.463 %	1.465 %	Vals AI Source details Data as of: 10/12/2026
CyberBench v1.1Patch Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Output limit: 262,144 tokens	No direct winner82.143 %	No direct winner82.143 %	Vals AI Source details Data as of: 10/12/2026
EMBComps Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value77.953 %	70.837 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Adjustments Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value48.318 %	45.274 %	Vals AI Source details Data as of: 10/12/2026
Harvey's Legal Agent BenchmarkAntitrust Competition · Criteria passed Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: criteria pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	No direct winner88.768 %	No direct winner88.768 %	Vals AI Source details Data as of: 10/12/2026
IOIIOI 2024 Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value100 %	59.667 %	Vals AI Source details Data as of: 10/14/2026
Legal Research BenchAdministrative / Regulatory · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	53.03 %	Higher measured value63.636 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevImage to webdev Test details No documented difference in the test setupVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDev Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevGPT-6 Astra: Arena rating from 2,739 published pairwise votes.Kimi K3: Arena rating from 3,504 published pairwise votes.	Winner1,731.395 95% confidence interval: ±13.296The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,579.165 95% confidence interval: ±10.818The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
LMArena Text, Style ControlBusiness, management and finance Test details No documented difference in the test setupNo clear leadVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 2,061 published pairwise votes.Kimi K3: Arena rating from 6,035 published pairwise votes.	1,472.34 95% confidence interval: ±13.336The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	No clear lead1,487.914 95% confidence interval: ±8.167The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
ProgramBenchAlmost Resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: almost-resolved rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value50 %	17.5 %	Vals AI Source details Data as of: 10/12/2026
SAGEAutomata Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	37.302 %	Higher measured value43.254 %	Vals AI Source details Data as of: 10/08/2026
SWE-bench Multimodalbpmn-js Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value62.264 %	47.17 %	Vals AI Source details Data as of: 10/13/2026
Tax Agent BenchControversy & Precedence Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	No direct winner13.333 %	No direct winner13.333 %	Vals AI Source details Data as of: 10/12/2026
Terminal-Bench 2.1Easy Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value100 %	91.667 %	Vals AI Source details Data as of: 09/28/2026
Terminal-Bench 4.0Hardware Test details Test setup not documented as identicalNo direct winnerVersion: 4.0Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value60 %	13.333 %	Vals AI Source details Data as of: 10/13/2026
Terminal-Bench ScienceEarth Sciences Test details Test setup not documented as identicalNo direct winnerVersion: 0.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: Mini-SWE-agent, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Harness: Mini-SWE-agent, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value62.5 %	12.5 %	Vals AI Source details Data as of: 10/14/2026
Time Horizon Index: KSPProgress efficiency Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: progressScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value70.112 %	9.258 %	Vals AI Source details Data as of: 10/07/2026
Vals IndexCode Migration Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: benchmark scoreScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value68.462 %	16.042 %	Vals AI Source details Data as of: 10/13/2026
BioMysteryBenchHuman-solvable Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: Terminus 2, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Harness: Terminus 2, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value89.041 %	79.909 %	Vals AI Source details Data as of: 10/12/2026
Code MigrationCOBOL Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value68.571 %	60 %	Vals AI Source details Data as of: 10/12/2026
CyberBench v1.1PoC Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Output limit: 262,144 tokens	0 %	Higher measured value68.333 %	Vals AI Source details Data as of: 10/12/2026
EMBDataroom Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	79.319 %	Higher measured value82.661 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)All-Pass Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	40.252 %	Higher measured value41.837 %	Vals AI Source details Data as of: 10/12/2026
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 betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	0 %	Higher measured value20 %	Vals AI Source details Data as of: 10/12/2026
IOIIOI 2025 Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value100 %	29.333 %	Vals AI Source details Data as of: 10/14/2026
Legal Research BenchAdministrative / Regulatory · Weighted partial credit Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: weighted partial-credit scoreScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	87.482 %	Higher measured value89.252 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevWebdev Test details No documented difference in the test setupVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDev Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevGPT-6 Astra: Arena rating from 5,746 published pairwise votes.Kimi K3: Arena rating from 18,644 published pairwise votes.	Winner1,780.244 95% confidence interval: ±10.541The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,655.295 95% confidence interval: ±6.069The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
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 betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 877 published pairwise votes.Kimi K3: Arena rating from 2,319 published pairwise votes.	1,514.537 95% confidence interval: ±19.541The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	No clear lead1,542.023 95% confidence interval: ±12.722The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
ProgramBenchFully Resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: fully resolved rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value5.5 %	2 %	Vals AI Source details Data as of: 10/12/2026
SAGECalculus Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	No direct winner79.938 %	No direct winner79.938 %	Vals AI Source details Data as of: 10/08/2026
SWE-bench MultimodalCarbon Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value24.06 %	21.053 %	Vals AI Source details Data as of: 10/13/2026
Tax Agent BenchControversy & Precedence Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	58.616 %	Higher measured value68.8 %	Vals AI Source details Data as of: 10/12/2026
Terminal-Bench 2.1Hard Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Temperature: 1, Output limit: 262,144 tokens	Higher measured value87.778 %	74.444 %	Vals AI Source details Data as of: 09/28/2026
Terminal-Bench 4.0Media Test details Test setup not documented as identicalNo direct winnerVersion: 4.0Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value66.667 %	16.667 %	Vals AI Source details Data as of: 10/13/2026
Terminal-Bench ScienceEngineering Sciences Test details Test setup not documented as identicalNo direct winnerVersion: 0.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: Mini-SWE-agent, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Harness: Mini-SWE-agent, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value33.333 %	0 %	Vals AI Source details Data as of: 10/14/2026
Vals IndexEMB Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: benchmark scoreScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value71.7 %	66.677 %	Vals AI Source details Data as of: 10/13/2026
Code MigrationCode Quality Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: mean reviewer code-quality ratingScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value4.469 / 5	3.899 / 5	Vals AI Source details Data as of: 10/12/2026
EMBDCF Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	Higher measured value64.483 %	56.736 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Comparables Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens	37.758 %	Higher measured value47.778 %	Vals AI Source details Data as of: 10/12/2026
Key to the marks:better valuebetter, but not conclusivetie

Our model tests

Our tests

SVG short test

One SVG prompt and fixed output settings. Captures are shown only with 0 reported reasoning tokens or documented lack of reasoning controls. Identical test conditions do not make the models' internal compute identical.

Shared test round: 4ae5d8eeb9a9

GPT-6 Astra

Unavailable in this round

Kimi K3

Earlier tests

Pelican on a bicycle

A community classic for free-form SVG drawing.

GPT-6 Astra

Reasoning (requested): high

Current attempt

Run date: Sep 7, 2026
API provider: openrouter
Requested API model: openai/gpt-6-astra
API model reported in response: openai/gpt-6-astra
Reasoning setting: high
Pinned endpoint: OpenAI | openai/gpt-6-astra-20260903
Temperature: Provider default
Token limit including reasoning: 8,192 tokens
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating
Response time: 141.7 s

Visually reviewed. This individual test does not establish overall model quality.

Kimi K3

Reasoning (requested): high

Current attempt

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: moonshotai/kimi-k3
API model reported in response: moonshotai/kimi-k3
Reasoning setting: high
Pinned endpoint: Moonshot AI | moonshotai/kimi-k3-20260715
Temperature: Provider default
Token limit including reasoning: 8,192 tokens
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating
Response time: 173.4 s

Visually reviewed. This individual test does not establish overall model quality.

Earlier attempts (1)

Sep 7, 2026API request failedReasoning setting: high

Prompt and test conditions

Generate an SVG of a pelican riding a bicycle

Token limit including reasoning: 8,192 tokens

Task origin (Simon Willison)

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 GPT-6 Astra and Kimi K3 with other leading models using the same data and benchmark logic.

View the complete LLM comparison

GPT-6 Astra vs. Kimi K3

Change model selection

General specifications

API pricing compared

Shared benchmarks

Capability profile from matched benchmarks

Our tests

SVG short test

GPT-6 Astra

Kimi K3

Pelican on a bicycle

Winning one benchmark is not an overall verdict

Related LLM comparisons