Skip to main content

Head-to-head comparison

Claude Opus 4.6 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
13
distinct benchmark metrics
4
latest retrieval
2026-10-16

Change model selection

Specifications

General specifications

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

Specifications for Claude Opus 4.6 and Kimi K3
AttributeClaude Opus 4.6Kimi K3
ProviderAnthropicMoonshot AI
ReleasedFeb 2026Jul 16, 2026
AvailabilityActiveActive
Model typeProprietaryOpen Weights
ParametersNot published2.8T, 104B active
ArchitectureNot publishedMixture of Experts
Context window1,000,000 tokens1,048,576 tokens
Knowledge cutoffMay 2025Not published
Technical sourcesModel, ContextModel

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 Claude Opus 4.6 and Kimi K3
Price typeClaude Opus 4.6Kimi K3
API input$5$3
API output$25$15
Cache read$0.5$0.3
Cache write (5 min.)$6.25Not listed
Cache write (1 hr.)$10Not listed
Price verified09/22/2026 Source09/22/2026 Source
Claude Opus 4.6Kimi K3
API inputUSD per 1M tokens, base tier
Claude Opus 4.6$5
Kimi K3Lower price$3
API outputUSD per 1M tokens, base tier
Claude Opus 4.6$25
Kimi K3Lower price$15
Cache readUSD per 1M tokens, base tier
Claude Opus 4.6$0.5
Kimi K3Lower price$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 Claude Opus 4.6 and Kimi K3
ModelProviderPurchase route and tariffInput per 1M tokensOutput per 1M tokensSource
Claude Opus 4.6AnthropicDirect from developerStandard APILowest listed price: $5Lowest listed price: $25Source09/22/2026
Amazon BedrockVia OpenRouteramazon-bedrockLowest listed price: $5Lowest listed price: $25Source10/26/2026
AnthropicVia OpenRouteranthropicLowest listed price: $5Lowest listed price: $25Source10/26/2026
AzureVia OpenRouterazure/globalLowest listed price: $5Lowest listed price: $25Source10/26/2026
Claude Platform on AWSVia OpenRouterclaude-on-awsLowest listed price: $5Lowest listed price: $25Source10/26/2026
GoogleVia OpenRoutergoogle-vertex/europe$5.5$27.5Source10/26/2026
GoogleVia OpenRoutergoogle-vertex/globalLowest listed price: $5Lowest listed price: $25Source10/26/2026
Kimi K3Moonshot AIDirect from developerStandard API$3$15Source09/22/2026
AkashMLVia OpenRouterakashml/fp4$1.2$14Source10/26/2026
AlibabaVia OpenRouteralibaba$3.45$17.25Source10/26/2026
Amazon BedrockVia OpenRouteramazon-bedrock/us-east-2$3$15Source10/26/2026
BaseTenVia OpenRouterbaseten/fp8$3$15Source10/26/2026
ChutesVia OpenRouterchutes/mxfp4$1.85$9.5Source10/26/2026
DecartVia OpenRouterdecart/mxfp4$1.65$8.25Source10/26/2026
DeepInfraVia OpenRouterdeepinfra/mxfp4$2.85$14.25Source10/26/2026
DigitalOceanVia OpenRouterdigitalocean$2.55$12.95Source10/26/2026
FireworksVia OpenRouterfireworks$3$15Source10/26/2026
FireworksVia OpenRouterfireworks/fast$4.5$22.5Source10/26/2026
FireworksVia OpenRouterfireworks/us$4.5$22.5Source10/26/2026
InferenceNetVia OpenRouterinference-net/fast$0.5$11.88Source10/26/2026
InferenceNetVia OpenRouterinference-net/fp4$0.239$10.8Source10/26/2026
InferenceNetVia OpenRouterinference-net/us$0.239$10.8Source10/26/2026
InferenceNetVia OpenRouterinference-net/us/fast$2.24$14.519Source10/26/2026
MakoraVia OpenRoutermakora/fp4$0.65$15Source10/26/2026
ModalVia OpenRoutermodal/mxfp4$3$15Source10/26/2026
Moonshot AIVia OpenRoutermoonshotai/mxfp4$3$15Source10/26/2026
MorphVia OpenRoutermorph/fp8$1.17$18.769Source10/26/2026
ParasailVia OpenRouterparasail/fp4$3$15Source10/26/2026
ParetoAlphaVia OpenRouterpareto-alpha/us$3$15Source10/26/2026
PhalaVia OpenRouterphala$2.55$12.75Source10/26/2026
RelaceVia OpenRouterrelace/fp4Lowest listed price: $0.2147$7.6972Source10/26/2026
Sail ResearchVia OpenRoutersail-research/fp4$1.2Lowest listed price: $5.5Source10/26/2026
Sail ResearchVia OpenRoutersail-research/fp4/us$1.2Lowest listed price: $5.5Source10/26/2026
WaferVia OpenRouterwafer$0.226$20Source10/26/2026
WaferVia OpenRouterwafer/us$0.36$11.9Source10/26/2026

Benchmark index

Benchmark index compared

The index combines independent benchmarks on a scale from 0 to 100. A lead only counts when a model comes first in at least 95% of 200 recalculations with randomly redrawn test families. Otherwise both count as level.

Open the AI leaderboard
Claude Opus 4.6

Overall index

No overall score yet

Kimi K3

Overall index

62.0

Kimi K3 leads in 1 of 5 shared areas. 4 areas are tied.

AreaClaude Opus 4.6Kimi K3
Codingtied43.852.2
Agents48.061.7
Financetied72.970.3
Legaltied71.573.4
Healthtied63.068.9

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.

Not enough directly matched categories for a reliable radar

Only 0 of at least 5 categories have comparable, direct model measurements. The benchmark tables below show the available evidence without filling missing axes.

Claude Opus 4.6Kimi K3
LMArena Code Arena WebDevOverall. Higher is better. Test setup not documented as identical
Claude Opus 4.61,537.258
Kimi K3Higher measured value1,655.295
MedCodeOverall. Higher is better. Test setup not documented as identical
Claude Opus 4.648.244 %
Kimi K3Higher measured value49.357 %
MedScribeOverall. Higher is better. Test setup not documented as identical
Claude Opus 4.686.738 %
Kimi K3Higher measured value88.048 %
Vibe Code Bench v1.1Overall. Higher is better. Test setup not documented as identical
Claude Opus 4.657.573 %
Kimi K3Higher measured value84.968 %
LMArena Code Arena WebDevOverall. Higher is better. Test setup not documented as identical
Claude Opus 4.61,546.362
Kimi K3Higher measured value1,655.295
LMArena Code Arena WebDevImage to webdev. Higher is better. Test setup not documented as identical
Claude Opus 4.61,538.785
Kimi K3Higher measured value1,579.165
LMArena Code Arena WebDevImage to webdev. Higher is better. Test setup not documented as identical
Claude Opus 4.61,542.687
Kimi K3Higher measured value1,579.165
LMArena Code Arena WebDevWebdev. Higher is better. Test setup not documented as identical
Claude Opus 4.61,537.258
Kimi K3Higher measured value1,655.295
LMArena Code Arena WebDevWebdev. Higher is better. Test setup not documented as identical
Claude Opus 4.61,546.362
Kimi K3Higher measured value1,655.295
LMArena Code Arena WebDevWebdev-html. Higher is better. Test setup not documented as identical
Claude Opus 4.61,546.218
Kimi K3Higher measured value1,645.8
LMArena Code Arena WebDevWebdev-html. Higher is better. Test setup not documented as identical
Claude Opus 4.61,558.715
Kimi K3Higher measured value1,645.8
LMArena Code Arena WebDevWebdev-react. Higher is better. Test setup not documented as identical
Claude Opus 4.61,528.416
Kimi K3Higher measured value1,670.362

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for Claude Opus 4.6 and Kimi K3
Benchmark and taskClaude Opus 4.6Kimi K3Source
LMArena Code Arena WebDevOverall
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 20,698 published pairwise votes.Kimi K3: Arena rating from 18,644 published pairwise votes.
1,537.258
95% confidence interval: ±5.345The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,655.295
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
MedCodeOverall
Test details
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.6: Compute: max, Temperature: 1, Output limit: 30,000 tokens
Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens
48.244 %
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 betterClaude Opus 4.6: Compute: max, Temperature: 1, Output limit: 30,000 tokens
Kimi K3: Reasoning: max, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens
86.738 %
Higher measured value88.048 %
Vals AI
Source details
Data as of: 10/13/2026
Vibe Code Bench v1.1Overall
Test details
Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterClaude Opus 4.6: Compute: max, Harness: OpenHands, Temperature: 1, Output limit: 128,000 tokens
Kimi K3: Reasoning: max, Harness: OpenHands, Temperature: 1, Top P: 0.95, Output limit: 262,144 tokens
57.573 %
Higher measured value84.968 %
Vals AI
Source details
Data as of: 10/12/2026
LMArena Code Arena WebDevOverall
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 19,443 published pairwise votes.Kimi K3: Arena rating from 18,644 published pairwise votes.
1,546.362
95% confidence interval: ±5.46The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,655.295
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 Code Arena WebDevImage to webdev
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 8,068 published pairwise votes.Kimi K3: Arena rating from 3,504 published pairwise votes.
1,538.785
95% confidence interval: ±9.009The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,579.165
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 Code Arena WebDevImage to webdev
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 7,930 published pairwise votes.Kimi K3: Arena rating from 3,504 published pairwise votes.
1,542.687
95% confidence interval: ±8.913The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,579.165
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 Code Arena WebDevWebdev
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 20,698 published pairwise votes.Kimi K3: Arena rating from 18,644 published pairwise votes.
1,537.258
95% confidence interval: ±5.345The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,655.295
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 Code Arena WebDevWebdev
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 19,443 published pairwise votes.Kimi K3: Arena rating from 18,644 published pairwise votes.
1,546.362
95% confidence interval: ±5.46The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,655.295
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 Code Arena WebDevWebdev-html
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 2,623 published pairwise votes.Kimi K3: Arena rating from 2,538 published pairwise votes.
1,546.218
95% confidence interval: ±12.521The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,645.8
95% confidence interval: ±13.981The 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 Code Arena WebDevWebdev-html
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 2,474 published pairwise votes.Kimi K3: Arena rating from 2,538 published pairwise votes.
1,558.715
95% confidence interval: ±13.217The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,645.8
95% confidence interval: ±13.981The 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 Code Arena WebDevWebdev-react
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 16,319 published pairwise votes.Kimi K3: Arena rating from 12,408 published pairwise votes.
1,528.416
95% confidence interval: ±5.908The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,670.362
95% confidence interval: ±7.159The 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 Code Arena WebDevWebdev-react
Test details
Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Code Arena WebDev
Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 15,177 published pairwise votes.Kimi K3: Arena rating from 12,408 published pairwise votes.
1,538.872
95% confidence interval: ±6.058The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
Higher measured value1,670.362
95% confidence interval: ±7.159The 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

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: 43b8c5c46830

Claude Opus 4.6

Claude Opus 4.6: SVG short test
API reports 0 reasoning tokensTest run: Oct 14, 2026Provider: anthropicAPI model: anthropic/claude-opus-4.6766 output tokens

Kimi K3

Kimi K3: SVG short test
API reports 0 reasoning tokensTest run: Oct 14, 2026Provider: chutes/mxfp4API model: moonshotai/kimi-k3495 output tokens
Earlier tests

Pelican on a bicycle

A community classic for free-form SVG drawing.

Claude Opus 4.6

Claude Opus 4.6: Pelican on a bicycle
A community classic for free-form SVG drawing.

Reasoning (requested): high

Current attempt
Run date
Oct 1, 2026
API provider
openrouter
Requested API model
anthropic/claude-opus-4.6
API model reported in response
anthropic/claude-opus-4.6
Reasoning setting
high
Pinned endpoint
Anthropic | anthropic/claude-4.6-opus-20260205
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
85.1 s

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

Kimi K3

Kimi K3: Pelican on a bicycle
A community classic for free-form SVG drawing.

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

Related LLM comparisons

Compare Claude Opus 4.6 and Kimi K3 with other leading models using the same data and benchmark logic.

View the complete LLM comparison