Compare published benchmark results from matching versions and cohorts, API pricing, and technical specifications. Any documented test setup differences remain visible.
shared result series
61
distinct benchmarks
3
latest retrieval
2026-09-29
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.
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.
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 DeepSeek-V4-Flash and Kimi K3
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 1 of at least 5 categories have comparable, direct model measurements. The benchmark tables below show the available evidence without filling missing axes.
DeepSeek-V4-FlashKimi K3
AlmanBench v0.1Accepted cases. Higher is better. No documented setup difference
DeepSeek-V4-FlashWinner93.003 %
Kimi K390.962 %
CyberBench v1.1Patch. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash72.414 %
Kimi K3Higher measured value83.051 %
LMArena Text, Style ControlOverall. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,436.082
Kimi K3Higher measured value1,487.964
CyberBench v1.1PoC. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash61.017 %
Kimi K3Higher measured value75 %
LMArena Text, Style ControlOverall. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,438.511
Kimi K3Higher measured value1,487.964
LMArena Text, Style ControlBusiness, management and finance. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,438.526
Kimi K3Higher measured value1,489.173
LMArena Text, Style ControlBusiness, management and finance. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,434.189
Kimi K3Higher measured value1,489.173
LMArena Text, Style ControlChinese. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,471.267
Kimi K3Higher measured value1,539.78
LMArena Text, Style ControlChinese. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,479.924
Kimi K3Higher measured value1,539.78
LMArena Text, Style ControlCoding. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,484.094
Kimi K3Higher measured value1,540.544
LMArena Text, Style ControlCoding. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,479.555
Kimi K3Higher measured value1,540.544
LMArena Text, Style ControlCreative writing. Higher is better. Test setup not documented as identical
DeepSeek-V4-Flash1,407.791
Kimi K3Higher measured value1,458.257
Key to the marks:better valuebetter, but not conclusivetie
Benchmark scores for DeepSeek-V4-Flash and Kimi K3
Benchmark and task
DeepSeek-V4-Flash
Kimi K3
Source
AlmanBench v0.1Accepted casesTest details
No documented difference in the test setupVersion: v0.1, results commit f34ddbfMetric: accepted casesScoring: Higher is betterSettings: Reasoning: max1,029 public tasks with one direct run per row. The target is the Alman language specification.
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 51,871 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 26,400 anonymous pairwise votes.
1,436.082
95% confidence interval: ±4.048The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,487.964
95% confidence interval: ±4.866The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterKimi K3: Temperature: 1, Output limit: 262,144 tokens
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 51,747 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 26,400 anonymous pairwise votes.
1,438.511
95% confidence interval: ±4.115The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,487.964
95% confidence interval: ±4.866The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlBusiness, management and financeTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 10,436 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 5,149 anonymous pairwise votes.
1,438.526
95% confidence interval: ±6.971The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,489.173
95% confidence interval: ±8.766The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlBusiness, management and financeTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 10,641 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 5,149 anonymous pairwise votes.
1,434.189
95% confidence interval: ±6.982The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,489.173
95% confidence interval: ±8.766The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 2,996 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 1,940 anonymous pairwise votes.
1,471.267
95% confidence interval: ±11.424The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,539.78
95% confidence interval: ±13.819The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 2,992 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 1,940 anonymous pairwise votes.
1,479.924
95% confidence interval: ±11.42The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,539.78
95% confidence interval: ±13.819The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 15,262 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 6,878 anonymous pairwise votes.
1,484.094
95% confidence interval: ±6.23The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,540.544
95% confidence interval: ±7.811The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 15,239 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 6,878 anonymous pairwise votes.
1,479.555
95% confidence interval: ±6.265The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,540.544
95% confidence interval: ±7.811The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 8,539 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 5,249 anonymous pairwise votes.
1,407.791
95% confidence interval: ±7.543The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,458.257
95% confidence interval: ±9.02The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 8,638 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 5,249 anonymous pairwise votes.
1,410.703
95% confidence interval: ±7.575The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,458.257
95% confidence interval: ±9.02The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 23,008 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 10,028 anonymous pairwise votes.
1,444.92
95% confidence interval: ±5.29The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,493.173
95% confidence interval: ±6.62The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 22,908 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 10,028 anonymous pairwise votes.
1,449.94
95% confidence interval: ±5.323The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,493.173
95% confidence interval: ±6.62The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlEntertainment, sports and mediaTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 11,179 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 6,417 anonymous pairwise votes.
1,404.567
95% confidence interval: ±6.937The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,459.187
95% confidence interval: ±8.256The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlEntertainment, sports and mediaTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 11,428 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 6,417 anonymous pairwise votes.
1,408.063
95% confidence interval: ±6.896The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,459.187
95% confidence interval: ±8.256The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 39,226 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 19,504 anonymous pairwise votes.
1,431.809
95% confidence interval: ±5.308The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,501.41
95% confidence interval: ±6.35The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 39,391 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 19,504 anonymous pairwise votes.
1,436.008
95% confidence interval: ±5.353The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,501.41
95% confidence interval: ±6.35The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 5,392 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 2,457 anonymous pairwise votes.
1,457.264
95% confidence interval: ±8.778The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,532.809
95% confidence interval: ±12.276The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 5,311 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 2,457 anonymous pairwise votes.
1,466.306
95% confidence interval: ±8.867The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,532.809
95% confidence interval: ±12.276The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 2,006 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 944 anonymous pairwise votes.
1,448.483
95% confidence interval: ±15.039The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,506.507
95% confidence interval: ±20.255The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 1,966 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 944 anonymous pairwise votes.
1,459.263
95% confidence interval: ±15.286The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,506.507
95% confidence interval: ±20.255The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 843 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 508 anonymous pairwise votes.
1,424.769
95% confidence interval: ±21.221The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,487.756
95% confidence interval: ±26.837The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 906 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 508 anonymous pairwise votes.
1,433.595
95% confidence interval: ±20.539The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,487.756
95% confidence interval: ±26.837The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 34,753 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 17,130 anonymous pairwise votes.
1,458.919
95% confidence interval: ±4.767The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,516.598
95% confidence interval: ±5.615The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 34,700 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 17,130 anonymous pairwise votes.
1,458.083
95% confidence interval: ±4.832The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,516.598
95% confidence interval: ±5.615The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 16,176 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 6,516 anonymous pairwise votes.
1,463.433
95% confidence interval: ±6.065The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,514.741
95% confidence interval: ±7.894The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 16,039 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 6,516 anonymous pairwise votes.
1,464.61
95% confidence interval: ±6.133The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,514.741
95% confidence interval: ±7.894The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 17,961 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 8,851 anonymous pairwise votes.
1,428.187
95% confidence interval: ±5.811The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,486.2
95% confidence interval: ±7.06The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 18,051 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 8,851 anonymous pairwise votes.
1,433.413
95% confidence interval: ±5.881The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,486.2
95% confidence interval: ±7.06The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 647 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 465 anonymous pairwise votes.
1,398.363
95% confidence interval: ±24.906The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,505.193
95% confidence interval: ±29.212The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 635 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 465 anonymous pairwise votes.
1,424.242
95% confidence interval: ±25.166The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,505.193
95% confidence interval: ±29.212The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 898 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 567 anonymous pairwise votes.
1,392.925
95% confidence interval: ±20.977The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,476.936
95% confidence interval: ±24.854The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 951 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 567 anonymous pairwise votes.
1,378.153
95% confidence interval: ±20.872The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,476.936
95% confidence interval: ±24.854The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlLegal and governmentTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 4,139 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 2,120 anonymous pairwise votes.
1,447.478
95% confidence interval: ±9.874The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,513.54
95% confidence interval: ±13.409The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlLegal and governmentTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 4,107 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 2,120 anonymous pairwise votes.
1,457.491
95% confidence interval: ±9.927The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,513.54
95% confidence interval: ±13.409The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlLife, physical and social scienceTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 8,423 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 4,134 anonymous pairwise votes.
1,456.345
95% confidence interval: ±7.388The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,521.44
95% confidence interval: ±9.547The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlLife, physical and social scienceTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 8,509 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 4,134 anonymous pairwise votes.
1,459.724
95% confidence interval: ±7.418The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,521.44
95% confidence interval: ±9.547The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 23,359 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 11,778 anonymous pairwise votes.
1,447.83
95% confidence interval: ±5.578The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,503.731
95% confidence interval: ±6.651The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 23,501 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 11,778 anonymous pairwise votes.
1,448.636
95% confidence interval: ±5.577The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,503.731
95% confidence interval: ±6.651The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 2,688 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 1,131 anonymous pairwise votes.
1,425.467
95% confidence interval: ±11.801The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,499.153
95% confidence interval: ±17.579The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 2,619 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 1,131 anonymous pairwise votes.
1,441.412
95% confidence interval: ±11.893The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,499.153
95% confidence interval: ±17.579The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 2,975 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 1,381 anonymous pairwise votes.
1,434.489
95% confidence interval: ±11.462The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,512.282
95% confidence interval: ±16.286The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 2,828 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 1,381 anonymous pairwise votes.
1,445.003
95% confidence interval: ±11.761The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,512.282
95% confidence interval: ±16.286The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlMedicine and healthcareTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 3,699 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 1,822 anonymous pairwise votes.
1,459.257
95% confidence interval: ±10.503The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,510.85
95% confidence interval: ±14.327The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlMedicine and healthcareTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 3,774 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 1,822 anonymous pairwise votes.
1,461.307
95% confidence interval: ±10.453The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,510.85
95% confidence interval: ±14.327The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 9,363 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 4,075 anonymous pairwise votes.
1,453.414
95% confidence interval: ±7.284The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,500.724
95% confidence interval: ±9.804The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 9,233 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 4,075 anonymous pairwise votes.
1,446.332
95% confidence interval: ±7.34The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,500.724
95% confidence interval: ±9.804The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 28,862 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 16,361 anonymous pairwise votes.
1,423.239
95% confidence interval: ±4.947The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,477.001
95% confidence interval: ±5.678The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 28,837 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 16,361 anonymous pairwise votes.
1,423.125
95% confidence interval: ±5.023The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,477.001
95% confidence interval: ±5.678The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 1,030 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 428 anonymous pairwise votes.
1,438.601
95% confidence interval: ±18.712The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,489.166
95% confidence interval: ±27.535The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 1,037 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 428 anonymous pairwise votes.
1,434.977
95% confidence interval: ±18.457The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,489.166
95% confidence interval: ±27.535The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 5,750 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 3,175 anonymous pairwise votes.
1,433.519
95% confidence interval: ±8.502The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,493.777
95% confidence interval: ±10.901The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 5,790 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 3,175 anonymous pairwise votes.
1,437.327
95% confidence interval: ±8.593The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,493.777
95% confidence interval: ±10.901The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlSoftware and IT servicesTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 21,360 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 10,253 anonymous pairwise votes.
1,470.864
95% confidence interval: ±5.508The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,526.074
95% confidence interval: ±6.653The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlSoftware and IT servicesTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 21,329 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 10,253 anonymous pairwise votes.
1,472.722
95% confidence interval: ±5.539The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,526.074
95% confidence interval: ±6.653The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 1,596 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 810 anonymous pairwise votes.
1,433.099
95% confidence interval: ±16.059The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,484.797
95% confidence interval: ±20.864The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 1,497 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 810 anonymous pairwise votes.
1,429.802
95% confidence interval: ±16.45The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,484.797
95% confidence interval: ±20.864The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlWriting, literature and languageTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 12,557 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 6,938 anonymous pairwise votes.
1,415.672
95% confidence interval: ±6.557The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,474.679
95% confidence interval: ±7.931The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
LMArena Text, Style ControlWriting, literature and languageTest details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterDeepSeek-V4-Flash: Snapshot: deepseek-v4-flash-high-preview, Reasoning: high, Harness: Arena text, style control Kimi K3: Snapshot: kimi-k3-max, Reasoning: max, Harness: Arena text, style controlDeepSeek-V4-Flash: Style-controlled Arena rating from 12,763 anonymous pairwise votes.Kimi K3: Style-controlled Arena rating from 6,938 anonymous pairwise votes.
1,422.277
95% confidence interval: ±6.549The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Higher measured value1,474.679
95% confidence interval: ±7.931The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
Key to the marks:better valuebetter, but not conclusivetie
Our model tests
Our tests
Compare how the models respond to the same prompt. Each test shows the first attempt, with no subsequent fixes to the generated code. These results do not contribute to an overall score.
Pelican on a bicycle
A community classic for free-form SVG drawing.
DeepSeek-V4-Flash
API access was unavailable for this run. The model was not evaluated.
Reasoning (requested): high
First attempt
Run date
7 Sept 2026
API provider
openrouter
Requested API model
deepseek/deepseek-v4-flash-0731
Reasoning setting
high
Pinned endpoint
DeepSeek | deepseek/deepseek-v4-flash-20260731
Temperature
Provider default
Token limit including reasoning
8,192 tokens
Test protocol
community-visual-tests-v1
Technical checks
Not run
Not yet visually reviewed. A successful recording does not confirm correct physics or full compliance with the prompt.
Kimi K3
API access was unavailable for this run. The model was not evaluated.
Reasoning (requested): high
First attempt
Run date
7 Sept 2026
API provider
openrouter
Requested API model
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
Not run
Not yet visually reviewed. A successful recording does not confirm correct physics or full compliance with the prompt.
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 DeepSeek-V4-Flash and Kimi K3 with other leading models using the same data and benchmark logic.