BenchLeader

Claude Opus 5.5 vs GPT-5.3-Codex

Full profiles: Claude Opus 5.5 (Anthropic) and GPT-5.3-Codex (OpenAI): every benchmark, price and rank.

Updated · Data from 12 sources

As of 12 Nov 2026, Claude Opus 5.5 is ahead of GPT-5.3-Codex on quality (68.5 vs 62.9 on the BenchLeader Index), but GPT-5.3-Codex is 1.7× cheaper at $4.81 vs $8.00 per 1M tokens blended. Choose Claude Opus 5.5 for coding (80 vs 61) and multimodal (70 vs 61), where it also streams ~1.5× faster; choose GPT-5.3-Codex for fast first answers (41.6 s vs 349.1 s) and agents & tools (80 vs 62).

Key facts (as of 12 Nov 2026)

  • Claude Opus 5.5 scores 68.5 on the BenchLeader Index and GPT-5.3-Codex scores 62.9 (BenchLeader, 12 Nov 2026).
  • Claude Opus 5.5 costs $8.00 per 1M tokens blended (list price, 12 Nov 2026).
  • GPT-5.3-Codex costs $4.81 per 1M tokens blended (list price, 12 Nov 2026).
  • Claude Opus 5.5 streams ~1.5× faster than GPT-5.3-Codex (Artificial Analysis, 7-day median, 12 Nov 2026).
  • Of the 6 benchmarks both models have, Claude Opus 5.5 leads 6 (BenchLeader, 12 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
CodingClaude Opus 5.580 vs 61 category scoreHigh
Agents & tool useGPT-5.3-Codex80 vs 62 category score; AA-Briefcase v1.1 868 vs 1801High
Long documentsEither67 vs 66 category score; AA-LCR v1.1 84.7% vs 83.3%Either
Images and other mediaClaude Opus 5.570 vs 61 category score; MMMU-Pro 87.7% vs 78.5%High
Lowest costGPT-5.3-Codex$4.81 vs $8.00 per 1M blended; agentic coding session $0.161 vs $0.320–
Fastest first answerGPT-5.3-Codex41.6 s (GPT-5.3-Codex (xhigh)) vs 349.1 s (Claude Opus 5.5 (max)) 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

  • Claude Opus 5.5 (max) leads on quality: 68.5 vs 62.9.
  • Claude Opus 5.5 (max) is stronger in coding, long context, multimodal.
  • GPT-5.3-Codex (xhigh) is stronger in agents & tools.
  • GPT-5.3-Codex (xhigh) is 1.7× cheaper ($4.81 vs $8.00 per 1M blended).
  • Claude Opus 5.5 (max) streams 1.5× faster (115 vs 78 tokens per second).

At a glance

SpecClaude Opus 5.5GPT-5.3-Codex
DeveloperAnthropicOpenAI
Released22 Sep 20265 Feb 2026
Context window (tokens)1,000,000400,000
Max output (tokens)128,000 (up to 300K on the Message Batches API (beta))–
Input typestext, image–
Reasoning effort settingslow, medium, high, xhigh, max–
Compared atClaude Opus 5.5 (max)GPT-5.3-Codex (xhigh)
Open weightsNoNo
Price in / out / cached input per 1M$4.00 / $20.00 / $0.200$1.75 / $14.00 / $0.175
API model IDclaude-opus-5-5–
Where to use itClaude API, Amazon Bedrock, Google Cloud, Microsoft Foundry–

Limits, availability and announced prices from the vendors' own pages: platform.claude.com (checked 20 Oct 2026). Everything else is from the day's data.

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 Opus 5.5 (max)GPT-5.3-Codex (xhigh)Leader
Reasoning72.3–not comparable
Coding79.961.0Claude Opus 5.5
Agents & tools61.580.1GPT-5.3-Codex
Maths73.2–not comparable
Knowledge72.8–not comparable
Instruction following–73.6not comparable
Multimodal70.360.9Claude Opus 5.5
Long context67.166.4level

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 avg75CodingAgents & toolsLong contextMultimodalClaude Opus 5.5, Coding: 79.9Claude Opus 5.5, Agents & tools: 61.5Claude Opus 5.5, Long context: 67.1Claude Opus 5.5, Multimodal: 70.3GPT-5.3-Codex, Coding: 61.0GPT-5.3-Codex, Agents & tools: 80.1GPT-5.3-Codex, Long context: 66.4GPT-5.3-Codex, Multimodal: 60.9

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

Model, click to add or removeIndex

Where sources disagree

Across the 6 benchmarks both have, Claude Opus 5.5 leads 6, GPT-5.3-Codex leads 0. Every overall index that rates both puts Claude Opus 5.5 ahead.

Overall indexClaude Opus 5.5 (max)GPT-5.3-Codex (xhigh)Leader
BenchLeader Index68.562.9Claude Opus 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 Opus 5.5 (max)GPT-5.3-Codex (xhigh)
Blended price per 1M tokens$8.00$4.81
Input / output per 1M$4.00 / $20.00$1.75 / $14.00
Cached input per 1M$0.200$0.175
Price per task: chat reply$0.0076$0.0049
Price per task: summarize a 30-page report$0.060$0.029
Price per task: code edit$0.054$0.032
Price per task: agentic coding session$0.320$0.161
Price per task: structured extraction$0.012$0.0063
Agentic coding session, input mostly cached$0.149$0.090
Cost per benchmark task (AA Index)$10.04–
Output tokens per task (AA Index)173k–
Of which reasoning120k–
Time per task (AA Index)25 min–
Output speed115 tok/s78 tok/s
Time to first answer349.1 s41.6 s
Time to a complete answer713.6 s90.3 s

Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 5.5 (max) scores 68.5 and takes 349 s to its first answer token; Claude Opus 5.5 (xhigh) scores 65.6 in 79 s, Claude Opus 5.5 (high) scores 65.4 in 22 s. GPT-5.3-Codex (xhigh) scores 62.9 and takes 42 s to its first answer token. 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

146 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 Opus 5.5 (max)GPT-5.3-Codex (xhigh)
BenchLeader IndexBenchLeader68.562.9
Agents & tools scoreBenchLeader61.580.1
Coding scoreBenchLeader79.961.0
Knowledge scoreBenchLeader72.8–
Long context scoreBenchLeader67.166.4
Maths scoreBenchLeader73.2–
Multimodal scoreBenchLeader70.360.9
Reasoning scoreBenchLeader72.3–
Instruction following scoreBenchLeader–73.6
Blended price $/MBenchLeader$8.00$4.81
Output speedBenchLeader115 tok/s78 tok/s
Time to first answerBenchLeader349.1 s41.6 s
Context windowBenchLeader1,000,000 tokens400,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub90.6%–
SimpleBenchSimpleBench88.4%†–
Arena Hard PromptsArena (LMArena)1530†–
LiveBench 2.0 ReasoningLiveBench92.2%†–
LiveBench 1.0 ReasoningLiveBench92.2%–
GPQA Diamond (AA)Artificial Analysis–91.5%
Humanity's Last Exam (AA)Artificial Analysis61.4%42.5%
HLE-Diamond (AA)Artificial Analysis67.9%–
ARC-AGI-1ARC Prize97.5%–
ARC-AGI-2ARC Prize91.7%–
CritPtArtificial Analysis31.7%16.9%
CritPt v2 hard subset (AA)Artificial Analysis68.0%–
MysteryMechanismVals AI49.5%†–
Chess PuzzlesEpoch AI Benchmarking Hub67.0%–
Mystery Game PuzzlesEpoch AI Benchmarking Hub71.0%–
LMCAEpoch AI Benchmarking Hub68.2%–
DTBenchEpoch AI Benchmarking Hub98.9%–
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–74.8%†
SciCodeSciCode66.9%–
WeirdMLWeirdML–77.9%
FrontierCodeCognition54.6%†–
Arena CodingArena (LMArena)1546†–
Arena WebDevArena (LMArena)18141406†
LiveBench 2.0 CodingLiveBench87.1%†–
LiveBench 1.0 CodingLiveBench89.3%–
Show all 146 benchmarks
MetricPublisherClaude Opus 5.5 (max)GPT-5.3-Codex (xhigh)
Coding
LiveCodeBenchVals AI–87.3%
IOIVals AI95.1%†53.8%
SWE-bench (Vals)Vals AI–78.0%
Code MigrationVals AI66.7%†–
Code Migration v1.1Vals AI81.0%†–
ProgramBenchVals AI18.5%†–
SWE-bench MultimodalVals AI70.2%†–
Vibe Code Bench 1-100Vals AI30.4%†–
Vibe Code Bench v1.1Vals AI90.3%†61.8%
MirrorCodeEpoch AI Benchmarking Hub95.6%–
CursorBenchCursor57.8%–
ALE-BenchEpoch AI Benchmarking Hub2147.4†1655.2
FrontierSWEEpoch AI Benchmarking Hub62.3%–
Terminal-Bench 4.0 (AA)Artificial Analysis59.6%–
Startrise FrontendStartrise71.1%†–
GameBench 2 OverallDuelLab GameBench 287.6†–
GameBench 2 by reasoning settingDuelLab GameBench 294.9†–
Design Arena WebsiteDesign Arena1323†1158†
FrontierCode 1.1 (AA)Artificial Analysis52.2%–
FrontierCode 2.0 (AA)Artificial Analysis46.4%–
Agents & tools
Terminal-BenchTerminal-Bench65.8%78.4%†
Remote Labor IndexScale AI / CAIS21.3%†–
APEX-AgentsMercor73.5%–
Arena AgentArena (LMArena)19.5–
LiveBench Long AgenticLiveBench72.8%†–
LiveBench BrowserLiveBench88.4%†–
LiveBench Agentic CodingLiveBench71.7%–
Terminal-Bench HardArtificial Analysis–53.0%
τ²-Bench Telecom (AA)Artificial Analysis–86.0%
Terminal-Bench 2.1 (Vals)Vals AI87.6%†–
MCP AtlasScale AI SEAL85.4%†–
HiL-BenchScale AI SEAL59.0%†4.3%†
GDPval-AA v2.1, normalised scoreArtificial Analysis68.3%–
GDPval-AA v2.1Artificial Analysis1866–
ITBench SRE (AA)Artificial Analysis38.2%–
Analyst Agent (AA)Artificial Analysis56.3%–
CUA-benchVals AI14.0%–
CyberBenchVals AI55.4%†–
Finance Agent v2Vals AI58.6%†–
Finance Agent v2.1Vals AI67.8%†–
Harvey's Legal Agent Benchmark v1.1Vals AI9.2%†–
Harvey's Legal Agent Benchmark v1Vals AI3.8%†–
Legal Research BenchVals AI50.5%†–
Robo-benchVals AI14.6%–
SREBenchVals AI33.6%†–
Tax Agent BenchVals AI44.5%†–
Terminal-Bench 2.0 (Vals)Vals AI–64.0%†
Terminal-Bench 4.0 (Vals)Vals AI65.2%†–
Terminal-Bench ScienceVals AI47.1%†–
Time Horizon Index: KSPVals AI91.3%†–
Web Search IndexVals AI53.8%†–
Arena SearchArena (LMArena)1213†–
PostTrainBenchEpoch AI Benchmarking Hub43.8%–
METR Time HorizonsMETR–74.5%†
Vending-Bench 2Andon Labs9235.3†5940.1†
AutomationBenchArtificial Analysis69.5%–
GDP.pdf (AA)Artificial Analysis26.2%–
Harvey LAB-AA v1.1Harvey4.2%–
AA-Briefcase v1.1Artificial Analysis1801868
GDPxGDPx by SemiAnalysis12.0%–
Design Arena AgenticDesign Arena1331†1011†
Terminal-Bench-Science 0.1 (AA)Artificial Analysis59.0%–
CWE-Bench (AA)Artificial Analysis58.3%–
DeepSec-Bench (AA)Artificial Analysis26.9%–
CyberGym end-to-end (AA)Artificial Analysis0.8%–
FACTS SearchGoogle88.4%†–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub91.2%–
FrontierMath Tier 4Epoch AI Benchmarking Hub95.0%–
OTIS Mock AIMEEpoch AI Benchmarking Hub100.0%–
ProofBenchVals AI100.0%–
LiveBench MathematicsLiveBench97.1%–
Arena MathsArena (LMArena)1519†–
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub72.2%–
LiveBench 2.0 Data AnalysisLiveBench80.3%†–
LiveBench ResearchLiveBench88.9%†–
LiveBench 1.0 Data AnalysisLiveBench80.3%–
AA-Omniscience v1.1Artificial Analysis48.1–
PRBench LegalScale AI SEAL47.6%†–
BioMysteryBenchVals AI79.3%†–
ClaimBenchVals AI100.0%†–
Excel Modeling BenchmarkVals AI75.9%†–
Excel Modeling Benchmark v1.1Vals AI75.8%†–
MedCodeVals AI49.8%†–
MedScribeVals AI91.4%†–
Public Benefits Bench v1.1Vals AI70.6%†–
EBR-benchEpoch AI Benchmarking Hub71.4%–
GDP.pdfEpoch AI Benchmarking Hub30.6%–
AA-Omniscience v1.1: accuracyArtificial Analysis65.5%–
AA-Omniscience v1.1: non-hallucinationArtificial Analysis49.7%–
HealthBench Professional (AA)OpenAI59.9%–
Instruction following
LiveBench LanguageLiveBench86.3%–
LiveBench Instruction FollowingLiveBench65.7%–
IFBenchArtificial Analysis–75.4%
Arena Instruction FollowingArena (LMArena)1519†–
Human preference
Arena TextArena (LMArena)1506†–
EQ-Bench 4EQ-Bench1340†–
Arena Creative WritingArena (LMArena)1518†–
Arena Multi-turnArena (LMArena)1501†–
Arena Longer QueriesArena (LMArena)1520†–
Multimodal
Arena VisionArena (LMArena)1324†–
LiveBench DesignLiveBench83.3%†–
MMMU-ProArtificial Analysis87.7%78.5%
SAGEVals AI91.3%†–
Blueprint-Bench 2Epoch AI Benchmarking Hub51.2%†–
Code2Video-AAArtificial Analysis1536–
AA-3DArtificial Analysis1202–
Long context
AA-LCR v1.1Artificial Analysis84.7%83.3%
MLCRArtificial Analysis27.8%–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub168.5†–
LiveBench 2.0LiveBench83.4%†–
LiveBench 1.0LiveBench83.2%–
AA Intelligence Index v5.0Artificial Analysis59.1–
AA Intelligence Index v4.3.2Artificial Analysis57.632.5
AA Progress Index v1Artificial Analysis350301.3
Vals IndexVals AI70.1†–
Vals Finance IndexVals AI65.3%†–
Vals RSI IndexVals AI37.3%–
Vals Science IndexVals AI58.6%†–
Vals Router IndexVals AI64.2%†–
AA Cyber IndexArtificial Analysis28.7–
Safety & honesty
FORTRESSScale AI SEAL5.8%–

Data as of 2026-11-12. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261110-2acac640a228 (https://benchmarks.duellab.org/releases/gb2-20261110-2acac640a228/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Claude Opus 5.5 (max): 1 new result (PostTrainBench); 1 score revision.
  • Claude Opus 5.5 (max): 1 new result (MirrorCode); 1 score revision.
  • Claude Opus 5.5 (max): BenchLeader Index 69.3 → 68.8.
  • Claude Opus 5.5 (max): 2 score revisions.
  • Claude Opus 5.5 (max): 1 score revision.
  • Claude Opus 5.5 (max): 1 score revision.
  • Claude Opus 5.5 (max): 2 score revisions.
  • Claude Opus 5.5 (max): 2 score revisions.
  • Claude Opus 5.5 (max): 1 score revision.
  • GPT-5.3-Codex (xhigh): 1 new result (AA-Briefcase v1.1).

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

Claude Opus 5.5 vs GPT-5.3-Codex: questions

Is Claude Opus 5.5 better than GPT-5.3-Codex?
Claude Opus 5.5 (max) leads on quality: 68.5 vs 62.9. The BenchLeader Index combines every independent quality benchmark; Claude Opus 5.5 (max) is ahead overall as of 12 Nov 2026, but check the category scores for your use.
Claude Opus 5.5 vs GPT-5.3-Codex: which is cheaper, and will that change?
GPT-5.3-Codex is cheaper today: $4.81 against $8.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Claude Opus 5.5 or GPT-5.3-Codex better for coding?
Claude Opus 5.5 scores higher in coding (80 vs 61 on the category index, where 50 is average).
Which is faster, Claude Opus 5.5 or GPT-5.3-Codex?
Claude Opus 5.5 streams ~1.5× faster (115 vs 78 output tokens per second, Artificial Analysis 7-day median).
Is Claude Opus 5.5 or GPT-5.3-Codex better for agents and tool use?
GPT-5.3-Codex scores higher in agentic tasks (80 vs 62 on the category index, where 50 is average).
More questions (4)
Which is better for long documents?
They are level on long-context benchmarks (67 vs 66). Claude Opus 5.5 accepts more context: 1,000,000 against 400,000 tokens for GPT-5.3-Codex.
Which has lower latency?
GPT-5.3-Codex answers first: 41.6 s against 349.1 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 Opus 5.5 vs GPT-5.3-Codex?
Across the 6 benchmarks both have, Claude Opus 5.5 leads 6, GPT-5.3-Codex leads 0. Every overall index that rates both puts Claude Opus 5.5 ahead.
When was Claude Opus 5.5 released?
Claude Opus 5.5 was released on 22 Sep 2026.