Enter opens the highlighted result. Esc closes.

Claude Sonnet 5 vs GPT-5.4 Mini

Updated October 21, 2026. 79 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Claude Sonnet 5 Replace GPT-5.4 Mini
Claude Sonnet 5Anthropic
Coding
63.5
Overall
38.2
Output $/M
$10.00
77
GPT-5.4 MiniOpenAI
Coding
28.2
Overall
24.1
Output $/M
$4.50
1
vs boards won of 79 shared, 1 tied

Decision reading

Pick Claude Sonnet 5 for coding: it scores 63.5 on the Vibe Coding Index against 28.2 for GPT-5.4 Mini.

Pick GPT-5.4 Mini when output tokens cost 2.2x less, or when a solved task costs $0.70 against $1.87.

Model AClaude Sonnet 5Anthropic63.5/1002 independent benchmarks$1.87 per solved taskModel BGPT-5.4 MiniOpenAI28.2/1001 independent benchmark$0.70 per solved task
Shared results
79
Claude Sonnet 5 only
41
GPT-5.4 Mini only
9
Like-for-like categories
7 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Claude Sonnet 5 63.5 from 2 benchmarks, GPT-5.4 Mini 28.2 from 1 benchmark.
Widest category
Agentic work, 21.0 points apart on average over 5 shared benchmarks.
Widest single gap
ProofBench: Claude Sonnet 5 77.0% against GPT-5.4 Mini 21.0%, 56.0 points apart on a 0 to 100 scale.
Output price
Claude Sonnet 5 $10.00 against GPT-5.4 Mini $4.50 per million output tokens, 2.2x apart.
Context window
Claude Sonnet 5 takes 1M tokens, GPT-5.4 Mini 400K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryClaude Sonnet 5BarsGPT-5.4 MiniTies
Coding1601 tie
Intelligence20
Agentic work50
Reasoning120
Math60
Knowledge90
General271
All shared boards7711 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index160
Coding benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
LiveBench coding74.7%Not measuredOnly Claude Sonnet 5 measured
LMArena WebDev15401397Claude Sonnet 5
AA Coding Index71.556.1Claude Sonnet 5
LiveCodeBench (Vals AI)82.4%81.5%Claude Sonnet 5
SWE-bench Verified (Vals AI)79.6%73.0%Claude Sonnet 5
Terminal-Bench 4 (Vals AI)10.6%2.5%Claude Sonnet 5
Vibe Code Bench (Vals AI)81.3%48.0%Claude Sonnet 5
Code migration (Vals AI)44.4%12.9%Claude Sonnet 5
ProgramBench (Vals AI)0.0%0.0%Tie
IOI (Vals AI)45.0%Not measuredOnly Claude Sonnet 5 measured
Terminal-Bench 2.1 (Vals AI)74.5%54.7%Claude Sonnet 5
Vibe Code Bench 1-100 (Vals AI)13.8%Not measuredOnly Claude Sonnet 5 measured
SkillsBench (Vals AI)46.5%Not measuredOnly Claude Sonnet 5 measured
CyberBench (Vals AI)70.3%Not measuredOnly Claude Sonnet 5 measured
SciCode54.3%52.1%Claude Sonnet 5
Terminal-Bench Hard (Artificial Analysis run)Not measured52.3%Only GPT-5.4 Mini measured
FrontierCode42.7%27.0%Claude Sonnet 5
GSO-Bench37.3%Not measuredOnly Claude Sonnet 5 measured
WeirdML68.8%60.3%Claude Sonnet 5
ALE-Bench1463.11188.6Claude Sonnet 5
LMArena Text: coding14831438Claude Sonnet 5
LMArena Text: software and IT14741436Claude Sonnet 5
LMArena WebDev: React15411386Claude Sonnet 5
LMArena WebDev: HTML15271419Claude Sonnet 5
LMArena WebDev: from an image1540Not measuredOnly Claude Sonnet 5 measured
Neither model has a result on the other 13 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
AA Intelligence Index38.224.1Claude Sonnet 5
Epoch Capabilities Index156.9148.7Claude Sonnet 5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index50
Agentic work benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
AA Agentic Index43.617.9Claude Sonnet 5
Design Arena: agents building web apps1245Not measuredOnly Claude Sonnet 5 measured
Design Arena: agents building full-stack apps1249Not measuredOnly Claude Sonnet 5 measured
Design Arena: agents building mobile apps1215Not measuredOnly Claude Sonnet 5 measured
Design Arena: agents building Android apps1231Not measuredOnly Claude Sonnet 5 measured
Design Arena: agents building Godot games1225Not measuredOnly Claude Sonnet 5 measured
Design Arena: agents building games1227Not measuredOnly Claude Sonnet 5 measured
Design Arena: HTML slides1218Not measuredOnly Claude Sonnet 5 measured
Design Arena: PowerPoint slides1216Not measuredOnly Claude Sonnet 5 measured
Design Arena: spreadsheets1004Not measuredOnly Claude Sonnet 5 measured
LiveBench research82.8%Not measuredOnly Claude Sonnet 5 measured
LiveBench browser72.0%Not measuredOnly Claude Sonnet 5 measured
LiveBench long agentic39.0%Not measuredOnly Claude Sonnet 5 measured
SREBench (Vals AI)0.4%Not measuredOnly Claude Sonnet 5 measured
Finance Agent (Vals AI)53.9%45.4%Claude Sonnet 5
Tax Agent Bench (Vals AI)29.4%8.8%Claude Sonnet 5
Legal Research Bench (Vals AI)41.8%12.5%Claude Sonnet 5
Excel Modeling Benchmark (Vals AI)66.3%45.4%Claude Sonnet 5
Terminal-Bench-Science (Vals AI)5.7%Not measuredOnly Claude Sonnet 5 measured
τ²-Bench TelecomNot measured83.3%Only GPT-5.4 Mini measured
APEX-Agents54.5%Not measuredOnly Claude Sonnet 5 measured
DeepResearch BenchNot measured36.3%Only GPT-5.4 Mini measured
Vending-Bench 2$6,378Not measuredOnly Claude Sonnet 5 measured
Neither model has a result on the other 16 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index120
Reasoning benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
GPQA Diamond88.9%83.1%Claude Sonnet 5
LiveBench reasoning88.7%Not measuredOnly Claude Sonnet 5 measured
MMMU Pro (Vals AI)83.0%79.3%Claude Sonnet 5
GPQA Diamond (Artificial Analysis run)91.1%87.5%Claude Sonnet 5
Humanity's Last Exam (Artificial Analysis run)41.3%28.1%Claude Sonnet 5
GPQA Diamond (Epoch AI run)90.5%86.9%Claude Sonnet 5
ARC-AGINot measured63.7%Only GPT-5.4 Mini measured
ARC-AGI-2Not measured18.9%Only GPT-5.4 Mini measured
SimpleBench60.6%Not measuredOnly Claude Sonnet 5 measured
Chess puzzles45.0%23.0%Claude Sonnet 5
Mystery game puzzles35.0%11.0%Claude Sonnet 5
LMArena Text: hard prompts14611424Claude Sonnet 5
LMArena Text: expert14891435Claude Sonnet 5
LMArena Text: hard prompts in English14641420Claude Sonnet 5
LMArena Vision: diagrams12931266Claude Sonnet 5
LMArena Vision: homework13111285Claude Sonnet 5
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index60
Math benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
AIMENot measured95.6%Only GPT-5.4 Mini measured
OTIS Mock AIME94.7%88.9%Claude Sonnet 5
FrontierMath, tiers 1 to 365.6%51.2%Claude Sonnet 5
FrontierMath, tier 429.3%9.8%Claude Sonnet 5
ProofBench77.0%21.0%Claude Sonnet 5
LMArena Text: math14671418Claude Sonnet 5
LMArena Text: mathematical work14811429Claude Sonnet 5
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index90
Knowledge benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
MMLU Pro87.5%84.5%Claude Sonnet 5
LegalBench (Vals AI)83.9%Not measuredOnly Claude Sonnet 5 measured
CaseLaw (Vals AI)Not measured51.7%Only GPT-5.4 Mini measured
CorpFin (Vals AI)67.0%60.9%Claude Sonnet 5
TaxEval (Vals AI)75.6%71.2%Claude Sonnet 5
MortgageTax (Vals AI)70.0%63.5%Claude Sonnet 5
MedCode (Vals AI)47.5%Not measuredOnly Claude Sonnet 5 measured
SimpleQA Verified33.7%29.4%Claude Sonnet 5
LMArena Text: science14601417Claude Sonnet 5
LMArena Text: medicine and healthcare14421395Claude Sonnet 5
LMArena Text: legal and government14541414Claude Sonnet 5
LMArena Text: business and finance14361418Claude Sonnet 5
Neither model has a result on the other 5 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
Web design benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
Design Arena: websites1274Not measuredOnly Claude Sonnet 5 measured
Design Arena: UI components1279Not measuredOnly Claude Sonnet 5 measured
Design Arena: game dev1278Not measuredOnly Claude Sonnet 5 measured
Design Arena: data viz1251Not measuredOnly Claude Sonnet 5 measured
Design Arena: 3D1261Not measuredOnly Claude Sonnet 5 measured
Design Arena: all code categories1276Not measuredOnly Claude Sonnet 5 measured
Design Arena: SVG1193Not measuredOnly Claude Sonnet 5 measured
Design Arena: ASCII art1207Not measuredOnly Claude Sonnet 5 measured
LiveBench design82.0%Not measuredOnly Claude Sonnet 5 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index271
General benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
LiveBench data analysis71.7%Not measuredOnly Claude Sonnet 5 measured
Public Benefits Bench (Vals AI)66.0%Not measuredOnly Claude Sonnet 5 measured
SAGE (Vals AI)48.9%50.8%GPT-5.4 Mini
MedScribe (Vals AI)76.0%Not measuredOnly Claude Sonnet 5 measured
Vals Multimodal Index68.8%54.2%Claude Sonnet 5
IFBenchNot measured73.3%Only GPT-5.4 Mini measured
Long context reasoning (AA-LCR)82.0%77.0%Claude Sonnet 5
DTBench92.5%80.0%Claude Sonnet 5
LMCA50.0%40.8%Claude Sonnet 5
Surface Evolver Bench60.0%Not measuredOnly Claude Sonnet 5 measured
LMArena Text14421412Claude Sonnet 5
LMArena Text: instruction following14521405Claude Sonnet 5
LMArena Text: creative writing14161370Claude Sonnet 5
LMArena Vision12731245Claude Sonnet 5
LMArena Text: English14501412Claude Sonnet 5
LMArena Text: non-English14311405Claude Sonnet 5
LMArena Text: Chinese14781446Claude Sonnet 5
LMArena Text: German14421408Claude Sonnet 5
LMArena Text: Japanese14221375Claude Sonnet 5
LMArena Text: Korean14121370Claude Sonnet 5
LMArena Text: French14631441Claude Sonnet 5
LMArena Text: Spanish14381405Claude Sonnet 5
LMArena Text: Russian14501417Claude Sonnet 5
LMArena Text: Polish14251415Claude Sonnet 5
LMArena Text: multi-turn14531429Claude Sonnet 5
LMArena Text: longer queries14621408Claude Sonnet 5
LMArena Text: writing and language14361393Claude Sonnet 5
LMArena Text: entertainment and media14111373Claude Sonnet 5
LMArena Vision: English12751244Claude Sonnet 5
LMArena Vision: Chinese13311286Claude Sonnet 5
LMArena Vision: reading text12881253Claude Sonnet 5
LMArena Vision: humor12611235Claude Sonnet 5
LMArena Vision: recognizing thingsNot measured1218Only GPT-5.4 Mini measured
LMArena Vision: creative writing12461222Claude Sonnet 5
LMArena Document1466Not measuredOnly Claude Sonnet 5 measured
Neither model has a result on the other 11 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Sonnet 5 and GPT-5.4 Mini
BenchmarkClaude Sonnet 5GPT-5.4 MiniLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Claude Sonnet 5$0.0070Fits in one request
GPT-5.4 Mini$0.0030Fits in one request

GPT-5.4 Mini costs less

Repository review

50K input + 3K output tokens

Claude Sonnet 5$0.130Fits in one request
GPT-5.4 Mini$0.051Fits in one request

GPT-5.4 Mini costs less

Agent loop

200K input + 10K output tokens

Claude Sonnet 5$0.500Fits in one request
GPT-5.4 Mini$0.195Fits in one request

GPT-5.4 Mini costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Claude Sonnet 5 and GPT-5.4 Mini
SpecificationAnthropic Claude Sonnet 5OpenAI GPT-5.4 Mini
MakerAnthropicOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedJune 30, 2026March 17, 2026
Context window1M400K
Input $/Mtok$2.00$0.75
Output $/Mtok$10.00$4.50
Coding index63.528.2
Independent measurements behind it21
Cost per solved task$1.87$0.70
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Sonnet 5 better than GPT-5.4 Mini for coding?

Pick Claude Sonnet 5 for coding: it scores 63.5 on the Vibe Coding Index against 28.2 for GPT-5.4 Mini. Pick GPT-5.4 Mini when output tokens cost 2.2x less, or when a solved task costs $0.70 against $1.87.

Which is cheaper, Claude Sonnet 5 or GPT-5.4 Mini?

Claude Sonnet 5 costs $2.00 per million input tokens and $10.00 per million output tokens; GPT-5.4 Mini costs $0.75 and $4.50. Per solved SWE-bench Verified task, Claude Sonnet 5 costs $1.87 and GPT-5.4 Mini $0.70.

Which has the bigger context window, Claude Sonnet 5 or GPT-5.4 Mini?

Claude Sonnet 5, with 1M tokens against 400K.

How many benchmarks is the Claude Sonnet 5 vs GPT-5.4 Mini comparison based on?

79 benchmarks have a result for both models as of October 21, 2026. Claude Sonnet 5's coding index rests on 2 independent benchmarks and GPT-5.4 Mini's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next