Compare published benchmark results from matching versions and cohorts, API pricing, and technical specifications. Any documented test setup differences remain visible.
shared result series
140
distinct benchmarks
22
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.
Specifications for Claude Sonnet 5 and GPT-5.4 mini
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 Claude Sonnet 5 and GPT-5.4 mini
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.
Claude Sonnet 5
GPT-5.4 mini
Coding: 100 / 79.5 (6 benchmark families)
Reasoning: 100 / 95.8 (2 benchmark families)
Mathematics: 100 / 97.6 (1 benchmark family)
Knowledge: 100 / 97.1 (2 benchmark families)
Finance: 100 / 83.8 (7 benchmark families)
Legal: 100 / 42.8 (3 benchmark families)
Agentic tasks: 100 / 73.4 (1 benchmark family)
Multimodal: 100 / 95.5 (1 benchmark family)
51 matched rows from 21 benchmark families were considered. The radar shows 8 of 12 comparable categories.
Claude Sonnet 5GPT-5.4 mini
Code MigrationOverall. Higher is better. No documented setup difference
Claude Sonnet 5Winner44.392 %
GPT-5.4 mini12.936 %
CorpFin v2Overall. Higher is better. No documented setup difference
Claude Sonnet 5Winner66.977 %
GPT-5.4 mini60.917 %
EMBOverall. Higher is better. No documented setup difference
Claude Sonnet 5Winner66.316 %
GPT-5.4 mini45.426 %
Finance Agent (v2)Overall. Higher is better. No documented setup difference
Claude Sonnet 5Winner53.909 %
GPT-5.4 mini45.36 %
GPQA DiamondOverall. Higher is better. No documented setup difference
Claude Sonnet 5Winner88.889 %
GPT-5.4 mini83.08 %
Harvey's Legal Agent BenchmarkOverall · Task fully resolved. Higher is better. No documented setup difference
Claude Sonnet 5Winner5 %
GPT-5.4 mini0 %
Legal Research BenchOverall · All-pass. Higher is better. No documented setup difference
Claude Sonnet 5Winner41.827 %
GPT-5.4 mini12.5 %
LiveCodeBench v6Overall. Higher is better. No documented setup difference
Claude Sonnet 5Winner82.429 %
GPT-5.4 mini81.465 %
LMArena Text, Style ControlOverall. Higher is better. No documented setup difference
Claude Sonnet 5Winner1,461.526
GPT-5.4 mini1,447.017
MMLU-ProOverall. Higher is better. No documented setup difference
Claude Sonnet 5Winner87.545 %
GPT-5.4 mini84.554 %
MMMU-ProOverall. Higher is better. No documented setup difference
Claude Sonnet 5Winner83.006 %
GPT-5.4 mini79.249 %
MortgageTaxOverall. Higher is better. No documented setup difference
Claude Sonnet 5Winner70.032 %
GPT-5.4 mini63.514 %
Key to the marks:better valuebetter, but not conclusivetie
Benchmark scores for Claude Sonnet 5 and GPT-5.4 mini
No documented difference in the test setupVersion: 1Metric: task resolution rateScoring: Higher is betterGPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Sonnet 5: Snapshot: claude-sonnet-5-high, Reasoning: high, Harness: Arena text, style control GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena text, style controlClaude Sonnet 5: Style-controlled Arena rating from 43,340 anonymous pairwise votes.GPT-5.4 mini: Style-controlled Arena rating from 63,140 anonymous pairwise votes.
Winner1,461.526
95% confidence interval: ±4.283The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
1,447.017
95% confidence interval: ±3.76The 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: arxiv:2609.25804v2Metric: both-orders-correct, engineering/research macro averageScoring: Higher is betterClaude Sonnet 5: Reasoning: adaptive thinking, Output limit: 65,536 tokens, Attempts: 502 questions, two option orders each GPT-5.4 mini: Reasoning: high, Output limit: 65,536 tokens, Attempts: 502 questions, two option orders eachSame prompt for all models. Each of 502 questions is shown twice with the options reversed. Average of engineering and research success rates, requiring both orders to be correct. 65,536-token output budget; unparseable answers count as wrong.
Higher measured value51.6 %
0 of 1,004 responses were unparseable and counted as wrong. Not a Gradually test.
40.1 %
112 of 1,004 responses were unparseable and counted as wrong. Not a Gradually test.
No documented difference in the test setupVersion: 2Metric: weighted index scoreScoring: Higher is betterGPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17
No documented difference in the test setupVersion: 1.2Metric: weighted index scoreScoring: Higher is betterGPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17
No documented difference in the test setupVersion: 1.1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Harness: OpenHands GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Harness: OpenHands
LMArena Text, Style ControlBusiness, management and financeTest details
No documented difference in the test setupNo clear leadVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Sonnet 5: Snapshot: claude-sonnet-5-high, Reasoning: high, Harness: Arena text, style control GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena text, style controlClaude Sonnet 5: Style-controlled Arena rating from 8,542 anonymous pairwise votes.GPT-5.4 mini: Style-controlled Arena rating from 12,859 anonymous pairwise votes.
No clear lead1,464.812
95% confidence interval: ±7.251The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
1,459.736
95% confidence interval: ±6.499The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Sonnet 5: Snapshot: claude-sonnet-5-high, Reasoning: high, Harness: Arena text, style control GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena text, style controlClaude Sonnet 5: Style-controlled Arena rating from 3,360 anonymous pairwise votes.GPT-5.4 mini: Style-controlled Arena rating from 3,931 anonymous pairwise votes.
Winner1,495
95% confidence interval: ±10.767The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
1,473.944
95% confidence interval: ±10.215The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Sonnet 5: Snapshot: claude-sonnet-5-high, Reasoning: high, Harness: Arena text, style control GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena text, style controlClaude Sonnet 5: Style-controlled Arena rating from 11,904 anonymous pairwise votes.GPT-5.4 mini: Style-controlled Arena rating from 17,234 anonymous pairwise votes.
Winner1,519.26
95% confidence interval: ±6.461The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.
1,494.645
95% confidence interval: ±5.907The 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 latest attempt, with no subsequent fixes to the generated code. Earlier runs remain available in the test details. These results do not contribute to an overall score.
Pelican on a bicycle
A community classic for free-form SVG drawing.
Claude Sonnet 5
A community classic for free-form SVG drawing.
Reasoning (requested): high
Current attempt
Run date
7 Sept 2026
API provider
openrouter
Requested API model
anthropic/claude-sonnet-5
API model reported in response
anthropic/claude-sonnet-5
Reasoning setting
high
Pinned endpoint
Anthropic | anthropic/claude-sonnet-5-20260630
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
47.9 s
Visually reviewed. This individual test does not establish overall model quality.
GPT-5.4 mini
A community classic for free-form SVG drawing.
Reasoning (requested): high
Current attempt
Run date
2 Oct 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)
1 Oct 2026API request failedReasoning setting: high
7 Sept 2026API request failedReasoning setting: high
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 Sonnet 5 and GPT-5.4 mini with other leading models using the same data and benchmark logic.
