Skip to main content

Head-to-head comparison

Claude Opus 4.6 vs. GPT-5.4 mini

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

shared result series
27
distinct benchmark metrics
3
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 GPT-5.4 mini
AttributeClaude Opus 4.6GPT-5.4 mini
ProviderAnthropicOpenAI
ReleasedFeb 2026Mar 17, 2026
AvailabilityActiveActive
Model typeProprietaryProprietary
ParametersNot publishedNot published
ArchitectureNot publishedNot published
Context window1,000,000 tokens400,000 tokens
Knowledge cutoffMay 2025August 31, 2025
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 GPT-5.4 mini
Price typeClaude Opus 4.6GPT-5.4 mini
API input$5$0.75
API output$25$4.5
Cache read$0.5$0.075
Cache write (5 min.)$6.25Not listed
Cache write (1 hr.)$10Not listed
Price verified09/22/2026 Source09/22/2026 Source
Claude Opus 4.6GPT-5.4 mini
API inputUSD per 1M tokens, base tier
Claude Opus 4.6$5
GPT-5.4 miniLower price$0.75
API outputUSD per 1M tokens, base tier
Claude Opus 4.6$25
GPT-5.4 miniLower price$4.5
Cache readUSD per 1M tokens, base tier
Claude Opus 4.6$0.5
GPT-5.4 miniLower price$0.075

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 GPT-5.4 mini
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
GPT-5.4 miniOpenAIDirect from developerStandard API$0.75$4.5Source09/22/2026
AzureVia OpenRouterazure$0.75$4.5Source10/26/2026
AzureVia OpenRouterazure/us$0.825$4.95Source10/26/2026
OpenAIVia OpenRouteropenai$0.75$4.5Source10/26/2026
OpenAIVia OpenRouteropenai/fast$1.5$9Source10/26/2026
OpenAIVia OpenRouteropenai/flexLowest listed price: $0.375Lowest listed price: $2.25Source10/26/2026

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

GPT-5.4 mini

Overall index

3.9

Claude Opus 4.6 leads in 4 of 4 shared areas.

AreaClaude Opus 4.6GPT-5.4 mini
Coding43.85.9
Agents48.04.5
Finance72.93.6
Legal71.51.2

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

Claude Opus 4.6GPT-5.4 mini
LMArena Code Arena WebDevOverall. Higher is better. Test setup not documented as identical
Claude Opus 4.6Higher measured value1,537.258
GPT-5.4 mini1,397.568
LMArena Vision, Style ControlOverall. Higher is better. Test setup not documented as identical
Claude Opus 4.6Higher measured value1,294.46
GPT-5.4 mini1,250.938
Vibe Code Bench v1.1Overall. Higher is better. Test setup not documented as identical
Claude Opus 4.6Higher measured value57.573 %
GPT-5.4 mini47.969 %
LMArena Code Arena WebDevOverall. Higher is better. No documented setup difference
Claude Opus 4.6Winner1,546.362
GPT-5.4 mini1,397.568
LMArena Vision, Style ControlOverall. Higher is better. No documented setup difference
Claude Opus 4.6Winner1,299.232
GPT-5.4 mini1,250.938
LMArena Code Arena WebDevWebdev. Higher is better. Test setup not documented as identical
Claude Opus 4.6Higher measured value1,537.258
GPT-5.4 mini1,397.568
LMArena Vision, Style ControlChinese. Higher is better. Test setup not documented as identical
Claude Opus 4.6Higher measured value1,348.115
GPT-5.4 mini1,285.423
LMArena Code Arena WebDevWebdev. Higher is better. No documented setup difference
Claude Opus 4.6Winner1,546.362
GPT-5.4 mini1,397.568
LMArena Vision, Style ControlChinese. Higher is better. No documented setup difference
Claude Opus 4.6Winner1,351.953
GPT-5.4 mini1,285.423
LMArena Code Arena WebDevWebdev-html. Higher is better. Test setup not documented as identical
Claude Opus 4.6Higher measured value1,546.218
GPT-5.4 mini1,419.865
LMArena Vision, Style ControlCreative writing vision. Higher is better. Test setup not documented as identical
Claude Opus 4.6Higher measured value1,297.75
GPT-5.4 mini1,226.475
LMArena Code Arena WebDevWebdev-html. Higher is better. No documented setup difference
Claude Opus 4.6Winner1,558.715
GPT-5.4 mini1,419.865

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for Claude Opus 4.6 and GPT-5.4 mini
Benchmark and taskClaude Opus 4.6GPT-5.4 miniSource
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
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 20,698 published pairwise votes.GPT-5.4 mini: Arena rating from 11,978 published pairwise votes.
Higher measured value1,537.258
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.
1,397.568
95% confidence interval: ±6.522The 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 Vision, Style ControlOverall
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 26,853 published pairwise votes.GPT-5.4 mini: Arena rating from 25,733 published pairwise votes.
Higher measured value1,294.46
95% confidence interval: ±6.329The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,250.938
95% confidence interval: ±6.248The 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
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
GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Harness: OpenHands, Output limit: 128,000 tokens
Higher measured value57.573 %
47.969 %
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 betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Code Arena WebDev
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 19,443 published pairwise votes.GPT-5.4 mini: Arena rating from 11,978 published pairwise votes.
Winner1,546.362
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.
1,397.568
95% confidence interval: ±6.522The 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 Vision, Style ControlOverall
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 22,304 published pairwise votes.GPT-5.4 mini: Arena rating from 25,733 published pairwise votes.
Winner1,299.232
95% confidence interval: ±6.456The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,250.938
95% confidence interval: ±6.248The 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
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 20,698 published pairwise votes.GPT-5.4 mini: Arena rating from 11,978 published pairwise votes.
Higher measured value1,537.258
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.
1,397.568
95% confidence interval: ±6.522The 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 Vision, Style ControlChinese
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 1,685 published pairwise votes.GPT-5.4 mini: Arena rating from 1,662 published pairwise votes.
Higher measured value1,348.115
95% confidence interval: ±17.347The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,285.423
95% confidence interval: ±17.385The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.		LMArena
Source details
Data as of: 10/09/2026
LMArena Code Arena WebDevWebdev
Test details
No documented difference in the test setupVersion: 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
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 19,443 published pairwise votes.GPT-5.4 mini: Arena rating from 11,978 published pairwise votes.
Winner1,546.362
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.
1,397.568
95% confidence interval: ±6.522The 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 Vision, Style ControlChinese
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 1,488 published pairwise votes.GPT-5.4 mini: Arena rating from 1,662 published pairwise votes.
Winner1,351.953
95% confidence interval: ±17.967The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,285.423
95% confidence interval: ±17.385The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.		LMArena
Source details
Data as of: 10/09/2026
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
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 2,623 published pairwise votes.GPT-5.4 mini: Arena rating from 1,436 published pairwise votes.
Higher measured value1,546.218
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.
1,419.865
95% confidence interval: ±16.676The 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 Vision, Style ControlCreative writing vision
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 1,731 published pairwise votes.GPT-5.4 mini: Arena rating from 1,627 published pairwise votes.
Higher measured value1,297.75
95% confidence interval: ±15.034The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,226.475
95% confidence interval: ±15.662The 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-html
Test details
No documented difference in the test setupVersion: 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
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 2,474 published pairwise votes.GPT-5.4 mini: Arena rating from 1,436 published pairwise votes.
Winner1,558.715
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.
1,419.865
95% confidence interval: ±16.676The 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 Vision, Style ControlCreative writing vision
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 1,405 published pairwise votes.GPT-5.4 mini: Arena rating from 1,627 published pairwise votes.
Winner1,309.447
95% confidence interval: ±16.801The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,226.475
95% confidence interval: ±15.662The 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-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
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 16,319 published pairwise votes.GPT-5.4 mini: Arena rating from 9,859 published pairwise votes.
Higher measured value1,528.416
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.
1,386.539
95% confidence interval: ±7.117The 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 Vision, Style ControlDiagram
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 7,260 published pairwise votes.GPT-5.4 mini: Arena rating from 6,941 published pairwise votes.
Higher measured value1,331.371
95% confidence interval: ±8.825The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,283.959
95% confidence interval: ±8.767The 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-react
Test details
No documented difference in the test setupVersion: 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
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Code Arena WebDevClaude Opus 4.6: Arena rating from 15,177 published pairwise votes.GPT-5.4 mini: Arena rating from 9,859 published pairwise votes.
Winner1,538.872
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.
1,386.539
95% confidence interval: ±7.117The 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 Vision, Style ControlDiagram
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 5,999 published pairwise votes.GPT-5.4 mini: Arena rating from 6,941 published pairwise votes.
Winner1,324.675
95% confidence interval: ±9.238The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,283.959
95% confidence interval: ±8.767The 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 Vision, Style ControlEnglish
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 11,327 published pairwise votes.GPT-5.4 mini: Arena rating from 10,691 published pairwise votes.
Higher measured value1,296.058
95% confidence interval: ±8.463The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,248.605
95% confidence interval: ±8.435The 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 Vision, Style ControlEnglish
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 9,251 published pairwise votes.GPT-5.4 mini: Arena rating from 10,691 published pairwise votes.
Winner1,301.128
95% confidence interval: ±8.811The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,248.605
95% confidence interval: ±8.435The 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 Vision, Style ControlEntity recognition
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 190 published pairwise votes.GPT-5.4 mini: Arena rating from 184 published pairwise votes.
Higher measured value1,237.39
95% confidence interval: ±39.346The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,195.219
95% confidence interval: ±40.866The 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 Vision, Style ControlEntity recognition
Test details
No documented difference in the test setupNo clear leadVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 125 published pairwise votes.GPT-5.4 mini: Arena rating from 184 published pairwise votes.
No clear lead1,234.469
95% confidence interval: ±47.028The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,195.219
95% confidence interval: ±40.866The 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 Vision, Style ControlHomework
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 3,708 published pairwise votes.GPT-5.4 mini: Arena rating from 3,503 published pairwise votes.
Higher measured value1,311.92
95% confidence interval: ±10.944The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,291.223
95% confidence interval: ±11.062The 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 Vision, Style ControlHomework
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 2,849 published pairwise votes.GPT-5.4 mini: Arena rating from 3,503 published pairwise votes.
Winner1,324.17
95% confidence interval: ±12.006The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,291.223
95% confidence interval: ±11.062The 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 Vision, Style ControlHumor
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 1,025 published pairwise votes.GPT-5.4 mini: Arena rating from 869 published pairwise votes.
Higher measured value1,294.209
95% confidence interval: ±19.436The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,238.173
95% confidence interval: ±21.177The 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 Vision, Style ControlHumor
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlArena rating from 869 published pairwise votes.
Winner1,296.35
95% confidence interval: ±21.068The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,238.173
95% confidence interval: ±21.177The 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 Vision, Style ControlOcr
Test details
Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 19,427 published pairwise votes.GPT-5.4 mini: Arena rating from 18,276 published pairwise votes.
Higher measured value1,312.247
95% confidence interval: ±6.33The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,262.843
95% confidence interval: ±6.292The 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 Vision, Style ControlOcr
Test details
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 4.6: Snapshot: claude-opus-4-6-high, Reasoning: high, Harness: Arena Vision, style control
GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena Vision, style controlClaude Opus 4.6: Arena rating from 15,929 published pairwise votes.GPT-5.4 mini: Arena rating from 18,276 published pairwise votes.
Winner1,313.477
95% confidence interval: ±6.575The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.
1,262.843
95% confidence interval: ±6.292The 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

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

GPT-5.4 mini

Unavailable in this round

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.

GPT-5.4 mini

GPT-5.4 mini: Pelican on a bicycle
A community classic for free-form SVG drawing.

Reasoning (requested): high

Current attempt
Run date
Oct 2, 2026
API provider
openrouter
Requested API model
openai/gpt-5.4-mini
API model reported in response
openai/gpt-5.4-mini
Reasoning setting
high
Pinned endpoint
OpenAI | openai/gpt-5.4-mini-20260317
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
13.5 s

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

Earlier attempts (2)
  • Oct 1, 2026API request failedReasoning setting: high
  • 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 GPT-5.4 mini with other leading models using the same data and benchmark logic.

View the complete LLM comparison