BenchLeader

Claude Sonnet 5 vs GPT-5.5

Full profiles: Claude Sonnet 5 (Anthropic) and GPT-5.5 (OpenAI): every benchmark, price and rank.

Updated · Data from 13 sources

As of 16 Nov 2026, GPT-5.5 is ahead of Claude Sonnet 5 on quality (64.1 vs 59.0 on the BenchLeader Index), but Claude Sonnet 5 is 2.8× cheaper at $4.00 vs $11.25 per 1M tokens blended. GPT-5.5 leads in every category we measure both in. Claude Sonnet 5 streams faster.

Key facts (as of 16 Nov 2026)

  • Claude Sonnet 5 scores 59.0 on the BenchLeader Index and GPT-5.5 scores 64.1 (BenchLeader, 16 Nov 2026).
  • Claude Sonnet 5 costs $4.00 per 1M tokens blended (list price, 16 Nov 2026).
  • GPT-5.5 costs $11.25 per 1M tokens blended (list price, 16 Nov 2026).
  • Claude Sonnet 5 streams faster than GPT-5.5 (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 17 benchmarks both models have, GPT-5.5 leads 15 (BenchLeader, 16 Nov 2026).
  • On Arena WebDev, Claude Sonnet 5 is rated 1539 and GPT-5.5 1512 (LMArena, as published by 16 Nov 2026).

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingEither60 vs 61 category score; Arena WebDev 1539 vs 1512Either
Agents & tool useGPT-5.562 vs 53 category score; GDPval-AA v2.1 1357 vs 1262High
MathsGPT-5.569 vs 65 category scoreMedium
ReasoningGPT-5.568 vs 65 category score; Humanity's Last Exam (AA) 45.8% vs 35.7%Medium
Long documentsGPT-5.567 vs 63 category score; AA-LCR v1.1 84.3% vs 76.7%Medium
Images and other mediaGPT-5.562 vs 61 category scoreMedium
Chat quality (human preference)–Only Claude Sonnet 5 has an Arena rating (65 category score)Not comparable
Lowest costClaude Sonnet 5$4.00 vs $11.25 per 1M blended; agentic coding session $0.160 vs $0.420–
Fastest first answerClaude Sonnet 515.8 s (Claude Sonnet 5 (high)) vs 38.3 s (GPT-5.5 (xhigh)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • GPT-5.5 (xhigh) leads on quality: 64.1 vs 59.0.
  • GPT-5.5 (xhigh) is stronger in agents & tools, coding, long context, maths, multimodal, reasoning.
  • Claude Sonnet 5 (high) is 2.8× cheaper ($4.00 vs $11.25 per 1M blended).

At a glance

SpecClaude Sonnet 5GPT-5.5
DeveloperAnthropicOpenAI
Released30 Jun 202623 Apr 2026
Context window (tokens)1,000,0001,050,000
Compared atClaude Sonnet 5 (high)GPT-5.5 (xhigh)
Open weightsNoNo
Price in / out / cached input per 1M$2.00 / $10.00 / $0.200$5.00 / $30.00 / $0.500

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryClaude Sonnet 5 (high)GPT-5.5 (xhigh)Leader
Reasoning65.468.1GPT-5.5
Coding59.960.7level
Agents & tools52.961.8GPT-5.5
Maths65.369.3GPT-5.5
Knowledge–63.6not comparable
Instruction following–74.0not comparable
Human preference64.6–not comparable
Multimodal60.962.3GPT-5.5
Long context62.966.9GPT-5.5

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningCodingAgents & toolsMathsLong contextMultimodalClaude Sonnet 5, Reasoning: 65.4Claude Sonnet 5, Coding: 59.9Claude Sonnet 5, Agents & tools: 52.9Claude Sonnet 5, Maths: 65.3Claude Sonnet 5, Long context: 62.9Claude Sonnet 5, Multimodal: 60.9GPT-5.5, Reasoning: 68.1GPT-5.5, Coding: 60.7GPT-5.5, Agents & tools: 61.8GPT-5.5, Maths: 69.3GPT-5.5, Long context: 66.9GPT-5.5, Multimodal: 62.3

Not drawn, measured for fewer than two of these models: Knowledge, Instruction following, Human preference.

Model, click to add or removeIndex

Where sources disagree

Across the 17 benchmarks both have, Claude Sonnet 5 leads 2, GPT-5.5 leads 15. Every overall index that rates both puts GPT-5.5 ahead.

Overall indexClaude Sonnet 5 (high)GPT-5.5 (xhigh)Leader
BenchLeader Index59.064.1GPT-5.5
Arena Text14611481†GPT-5.5

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

MeasureClaude Sonnet 5 (high)GPT-5.5 (xhigh)
Blended price per 1M tokens$4.00$11.25
Input / output per 1M$2.00 / $10.00$5.00 / $30.00
Cached input per 1M$0.200$0.500
Price per task: chat reply$0.0038$0.011
Price per task: summarize a 30-page report$0.030$0.078
Price per task: code edit$0.027$0.075
Price per task: agentic coding session$0.160$0.420
Price per task: structured extraction$0.0060$0.016
Agentic coding session, input mostly cached$0.079$0.217
Output speed99 tok/s81 tok/s
Time to first answer15.8 s38.3 s
Time to a complete answer, last measured 13 Nov 202640.7 s104.3 s

Each model is compared at its best-ranked reasoning-effort setting. Claude Sonnet 5 (high) scores 59.0 and takes 16 s to its first answer token; Claude Sonnet 5 (max) scores 56.1 in 158 s, Claude Sonnet 5 (xhigh) scores 55.4 in 32 s. GPT-5.5 (xhigh) scores 64.1 and takes 38 s to its first answer token; GPT-5.5 (high) scores 63.5 in 16 s, GPT-5.5 scores 61.0 in 3 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

148 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherClaude Sonnet 5 (high)GPT-5.5 (xhigh)
BenchLeader IndexBenchLeader59.064.1
Agents & tools scoreBenchLeader52.961.8
Coding scoreBenchLeader59.960.7
Human preference scoreBenchLeader64.6–
Long context scoreBenchLeader62.966.9
Maths scoreBenchLeader65.369.3
Multimodal scoreBenchLeader60.962.3
Reasoning scoreBenchLeader65.468.1
Instruction following scoreBenchLeader–74.0
Knowledge scoreBenchLeader–63.6
Blended price $/MBenchLeader$4.00$11.25
Output speedBenchLeader99 tok/s81 tok/s
Time to first answerBenchLeader15.8 s38.3 s
Context windowBenchLeader1,000,000 tokens1,050,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub80.3%†90.7%†
SimpleBenchSimpleBench60.6%†69.0%†
Arena Hard PromptsArena (LMArena)14901499†
LiveBench 1.0 ReasoningLiveBench88.7%†89.7%
GPQA Diamond (AA)Artificial Analysis91.1%†93.5%
Humanity's Last Exam (AA)Artificial Analysis35.7%45.8%
GPQA Diamond (Vals)Vals AI88.9%†93.2%
Kagi LLM BenchmarkKagi LLM Benchmark–88.8%†
ARC-AGI-1ARC Prize–95.0%
ARC-AGI-2ARC Prize–85.0%
ARC-AGI-3ARC Prize–0.4%†
CritPtArtificial Analysis15.1%27.1%
Arena ExpertArena (LMArena)15091510†
Chess PuzzlesEpoch AI Benchmarking Hub45.0%†63.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub16.0%†56.0%
CL-bench LifeEpoch AI Benchmarking Hub–22.2%†
LMCAEpoch AI Benchmarking Hub50.0%54.3%
DTBenchEpoch AI Benchmarking Hub84.5%96.0%
ForecastBenchForecastBench63.1%†64.2%†
BTF-3Epoch AI Benchmarking Hub14.2%†14.3%†
Coding
SciCodeSciCode54.3%55.8%
WeirdMLWeirdML68.8%84.9%
FrontierCodeCognition42.7%†43.0%†
GSO-BenchGSO-Bench37.3%†40.2%
Arena CodingArena (LMArena)15201510†
Show all 148 benchmarks
MetricPublisherClaude Sonnet 5 (high)GPT-5.5 (xhigh)
Coding
Arena WebDevArena (LMArena)15391512
LiveBench 1.0 CodingLiveBench80.7%†82.2%
LiveCodeBenchVals AI82.4%†85.3%
IOIVals AI45.0%†–
SWE-bench (Vals)Vals AI79.6%†82.6%
Code MigrationVals AI44.4%†45.2%
ProgramBenchVals AI0.0%†0.5%
Vibe Code Bench 1-100Vals AI13.8%†–
Vibe Code Bench v1.1Vals AI81.3%†69.8%
SWE Atlas v1.0: Codebase QnAScale AI SEAL–45.4%
SWE Atlas v1.0: RefactoringScale AI SEAL–44.8%
SWE Atlas v1.0: Test WritingScale AI SEAL–42.6%
SWE-Bench Pro V2Scale AI SEAL93.2%†–
SWE-Bench Pro V2 HARDScale AI SEAL88.2%†–
MirrorCodeEpoch AI Benchmarking Hub63.3%†23.3%
Surface Evolver BenchEpoch AI Benchmarking Hub60.0%†88.1%†
DeepSWE v1.1Epoch AI Benchmarking Hub48.2%67.0%
CursorBenchCursor30.8%–
ALE-BenchEpoch AI Benchmarking Hub1463.11943.0
Terminal-Bench 4.0 (AA)Artificial Analysis5.0%14.7%
Terminal-Bench 2.1 (AA)Artificial Analysis80.5%†84.3%
Startrise FrontendStartrise66.6%†–
GameBench 2 OverallDuelLab GameBench 246.6†49.2†
GameBench 2 by reasoning settingDuelLab GameBench 254.4†59.9
Design Arena WebsiteDesign Arena1268†1248†
FrontierCode 2.0 (AA)Artificial Analysis17.8%†–
Agents & tools
Terminal-BenchTerminal-Bench12.1%†84.7%†
OSWorld-Verified 2.0OSWorld–13.0%
Remote Labor IndexScale AI / CAIS–6.3%†
APEX-AgentsMercor54.5%†55.1%†
Arena AgentArena (LMArena)2–
LiveBench Agentic CodingLiveBench59.4%†54.0%
Terminal-Bench HardArtificial Analysis–60.6%
τ²-Bench Telecom (AA)Artificial Analysis–93.9%
Terminal-Bench 2.1 (Vals)Vals AI74.5%†57.3%†
MCP AtlasScale AI SEAL–75.3%
HiL-BenchScale AI SEAL–39.7%†
τ³-bench Bankingτ²-bench–44.6%
GDPval-AA v2.1Artificial Analysis12621357
τ³-Banking (AA)Artificial Analysis37.3%†39.0%
ITBench SRE (AA)Artificial Analysis–45.8%
Analyst Agent (AA)Artificial Analysis46.3%†50.0%
APEX-Agents (AA)Artificial Analysis–37.7%
CyberBenchVals AI61.9%†–
Finance Agent v2Vals AI53.9%†51.8%
Harvey's Legal Agent Benchmark v1Vals AI5.0%†3.8%
Legal Research BenchVals AI41.8%†40.4%
SkillsBenchVals AI46.5%†62.2%
SREBenchVals AI0.4%†3.8%
Tax Agent BenchVals AI29.4%†21.8%
Terminal-Bench 2.0 (Vals)Vals AI–73.2%†
Terminal-Bench 4.0 (Vals)Vals AI10.6%†–
Terminal-Bench ScienceVals AI5.7%†–
Time Horizon Index: KSPVals AI–8.3%
Arena SearchArena (LMArena)1191†1226†
Arena Agent Task CompletionArena (LMArena)-0.6–
DeepResearch BenchDeepResearch Bench–54.0%†
PostTrainBenchEpoch AI Benchmarking Hub–23.9%
ExploitBenchEpoch AI Benchmarking Hub–47.4%†
GBAEvalEpoch AI Benchmarking Hub65.3%†53.2%†
Vending-Bench 2Andon Labs6377.7†7523.8†
AutomationBenchArtificial Analysis32.1%47.3%
GDP.pdf (AA)Artificial Analysis9.4%21.2%
EnterpriseOps-GymArtificial Analysis44.7%†46.6%
AA-Briefcase v1.1Artificial Analysis11771135
Design Arena AgenticDesign Arena1222†1113†
RoboDojo (LLM as robot policy)RoboDojo–1.1†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub65.6%†85.3%
FrontierMath Tier 4Epoch AI Benchmarking Hub29.3%†72.5%
OTIS Mock AIMEEpoch AI Benchmarking Hub80.0%†84.4%†
ProofBenchVals AI77.0%†50.0%
LiveBench MathematicsLiveBench92.9%†95.9%
AIME 2026MathArena–100.0%
HMMT February 2026MathArena–98.5%
USAMO 2026MathArena–98.2%
MathArena ApexMathArena–80.2%
Arena MathsArena (LMArena)14721500†
FrontierMath ErdősEpoch AI Benchmarking Hub–0.0%
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub33.7%†63.0%
LiveBench 1.0 Data AnalysisLiveBench71.7%†81.6%
MMLU-ProVals AI87.5%†88.1%
LegalBenchVals AI83.9%†86.5%
CorpFinVals AI67.0%†68.4%
TaxEvalVals AI75.6%†75.0%
CaseLaw v2Vals AI–66.2%
Excel Modeling BenchmarkVals AI66.3%†64.5%
MedCodeVals AI47.5%†49.1%
MedScribeVals AI76.0%†86.9%
Public Benefits Bench v1.1Vals AI66.0%†60.9%
DrugDiscoveryBenchScale AI SEAL50.0%†51.6%
EBR-benchEpoch AI Benchmarking Hub–34.3%
GDP.pdfEpoch AI Benchmarking Hub–26.0%
FACTS ParametricGoogle20.5%†–
Instruction following
LiveBench LanguageLiveBench75.0%†87.4%
LiveBench Instruction FollowingLiveBench63.9%†70.7%
IFBenchArtificial Analysis–75.8%
Arena Instruction FollowingArena (LMArena)14671475†
Human preference
Arena TextArena (LMArena)14611476†
EQ-Bench 4EQ-Bench1221†1295†
EQ-Bench Creative Writing v3EQ-Bench1798†1845†
Arena Creative WritingArena (LMArena)14341448†
Arena Multi-turnArena (LMArena)14731484†
Arena Longer QueriesArena (LMArena)14811486†
Multimodal
Arena VisionArena (LMArena)12721296†
MMMU-ProArtificial Analysis77.3%†79.9%
MMMU-Pro (Vals)Vals AI83.0%†88.3%
MortgageTaxVals AI70.0%†68.8%
Humanity's Sixth SenseScale AI SEAL24.8%†–
GeoBenchGeoBench74.0%–
Blueprint-Bench 2Epoch AI Benchmarking Hub24.9%†36.2%†
Furniture AssemblyEpoch AI Benchmarking Hub–44.2%
Code2Video-AAArtificial Analysis–1000
Long context
AA-LCR v1.1Artificial Analysis76.7%84.3%
Arena DocumentArena (LMArena)14691479†
MLCRArtificial Analysis4.4%28.9%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub156.9†160.1†
LiveBench 1.0LiveBench76.0%†80.2%
AA Intelligence Index v4.3.2Artificial Analysis31.738.4
AA Progress Index v1Artificial Analysis296.8320
Vals RSI IndexVals AI–16.5%
Vals Multimodal IndexVals AI68.8%†68.1%
Safety & honesty
FORTRESSScale AI SEAL–16.3%
DistressBenchScale AI SEAL68.0%†–

Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Claude Sonnet 5 (high): 2 new results (GDPval-AA v2.1, AA-Briefcase v1.1).
  • GPT-5.5 (xhigh): 1 new result (GDPval-AA v2.1).
  • GPT-5.5 (xhigh): BenchLeader Index 64.6 → 64.1.
  • Claude Sonnet 5 (high): BenchLeader Index 59.5 → 59.0.
  • GPT-5.5 (xhigh): 1 score revision.
  • Claude Sonnet 5 (high): 1 new result (GeoBench).
  • GPT-5.5 (xhigh): 1 new result (MirrorCode).
  • Claude Sonnet 5 (high): 1 score revision.
  • GPT-5.5 (xhigh): 1 score revision.
  • Claude Sonnet 5 (high): 1 score revision.

Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs

Claude Sonnet 5 vs GPT-5.5: questions

Is GPT-5.5 better than Claude Sonnet 5?
GPT-5.5 (xhigh) leads on quality: 64.1 vs 59.0. The BenchLeader Index combines every independent quality benchmark; GPT-5.5 (xhigh) is ahead overall as of 16 Nov 2026, but check the category scores for your use.
Claude Sonnet 5 vs GPT-5.5: which is cheaper, and will that change?
Claude Sonnet 5 is cheaper today: $4.00 against $11.25 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Claude Sonnet 5 or GPT-5.5 better for coding?
They are level in coding (60 vs 61 on the category index). On Arena WebDev, the top shared coding benchmark, it is 1539 for Claude Sonnet 5 against 1512 for GPT-5.5.
Which is faster, Claude Sonnet 5 or GPT-5.5?
Claude Sonnet 5 streams faster (99 vs 81 output tokens per second, Artificial Analysis 7-day median).
Is Claude Sonnet 5 or GPT-5.5 better for agents and tool use?
GPT-5.5 scores higher in agentic tasks (62 vs 53 on the category index, where 50 is average).
More questions (4)
Which is better for long documents?
GPT-5.5 scores higher on long-context benchmarks (67 vs 63). Both take about 1.1M tokens: Claude Sonnet 5 1,000,000, GPT-5.5 1,050,000.
Which has lower latency?
Claude Sonnet 5 answers first: 15.8 s against 38.3 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on Claude Sonnet 5 vs GPT-5.5?
Across the 17 benchmarks both have, Claude Sonnet 5 leads 2, GPT-5.5 leads 15. Every overall index that rates both puts GPT-5.5 ahead.
When was Claude Sonnet 5 released?
Claude Sonnet 5 was released on 30 Jun 2026.