BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases12

Claude Opus 4.8 vs GPT-6 Sol

Full profiles: Claude Opus 4.8 (Anthropic) and GPT-6 Sol (OpenAI): every benchmark, price and rank.

Updated · Data from 10 sources

As of 9 Nov 2026, Claude Opus 4.8 and GPT-6 Sol are neck and neck on quality (60.5 vs 62.9 on the BenchLeader Index, inside the ±6.4 margin), and GPT-6 Sol is 2.5× cheaper at $4.00 vs $10.00 per 1M tokens blended. Choose Claude Opus 4.8 for multimodal (64 vs 61) and human preference (67 vs 64) and fast first answers (0.8 s vs 72.1 s); choose GPT-6 Sol for reasoning (66 vs 60), coding (67 vs 61), and maths (69 vs 67), where it also streams ~1.5× faster.

Key facts (as of 9 Nov 2026)

  • Claude Opus 4.8 scores 60.5 on the BenchLeader Index and GPT-6 Sol scores 62.9 (BenchLeader, 9 Nov 2026).
  • Claude Opus 4.8 costs $10.00 per 1M tokens blended (list price, 9 Nov 2026).
  • GPT-6 Sol costs $4.00 per 1M tokens blended (list price, 9 Nov 2026).
  • GPT-6 Sol streams ~1.5× faster than Claude Opus 4.8 (Artificial Analysis, 7-day median, 9 Nov 2026).
  • Of the 13 benchmarks both models have, Claude Opus 4.8 leads 8 (BenchLeader, 9 Nov 2026).
  • On Arena WebDev, Claude Opus 4.8 is rated 1555 and GPT-6 Sol 1674 (LMArena, as published by 9 Nov 2026).
Claude Opus 4.8 benchmarks high Anthropic GPT-6 Sol benchmarks max OpenAIAdd a third model

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
CodingGPT-6 Sol67 vs 61 category score; Arena WebDev 1674 vs 1555Medium
MathsGPT-6 Sol69 vs 67 category score; Arena Maths 1440 vs 1492Medium
ReasoningGPT-6 Sol66 vs 60 category score; ARC-AGI-2 89.6% vs 72.1%Medium
Images and other mediaClaude Opus 4.864 vs 61 category score; Arena Vision 1295 vs 1232Medium
Chat quality (human preference)Claude Opus 4.867 vs 64 on Arena votesMedium
Lowest costGPT-6 Sol$4.00 vs $10.00 per 1M blended; agentic coding session $0.160 vs $0.400–
Fastest first answerClaude Opus 4.80.8 s (Claude Opus 4.8 (high)) vs 72.1 s (GPT-6 Sol (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 4.8 (high) and GPT-6 Sol (max) are neck and neck on quality (60.5 vs 62.9, within the ±6.4 margin).
  • Claude Opus 4.8 (high) is stronger in human preference, multimodal.
  • GPT-6 Sol (max) is stronger in coding, maths, reasoning, agents & tools, knowledge, long context.
  • GPT-6 Sol (max) is 2.5× cheaper ($4.00 vs $10.00 per 1M blended).
  • GPT-6 Sol (max) streams 1.7× faster (121 vs 71 tokens per second).

At a glance

SpecClaude Opus 4.8GPT-6 Sol
DeveloperAnthropicOpenAI
Released28 May 202622 Sep 2026
Context window (tokens)1,000,0001,050,000
Compared atClaude Opus 4.8 (high)GPT-6 Sol (max)
Open weightsNoNo
Price in / out / cached input per 1M$5.00 / $25.00 / $0.500$2.00 / $10.00 / $0.200

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 4.8 (high)GPT-6 Sol (max)Leader
Reasoning60.066.2GPT-6 Sol
Coding61.466.5GPT-6 Sol
Agents & tools–59.6not comparable
Maths67.468.5GPT-6 Sol
Knowledge–61.5not comparable
Human preference67.364.2Claude Opus 4.8
Multimodal64.060.7Claude Opus 4.8
Long context–66.6not comparable

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 avg75ReasoningCodingMathsMultimodalHumanpreferenceClaude Opus 4.8, Reasoning: 60.0Claude Opus 4.8, Coding: 61.4Claude Opus 4.8, Maths: 67.4Claude Opus 4.8, Multimodal: 64.0Claude Opus 4.8, Human preference: 67.3GPT-6 Sol, Reasoning: 66.2GPT-6 Sol, Coding: 66.5GPT-6 Sol, Maths: 68.5GPT-6 Sol, Multimodal: 60.7GPT-6 Sol, Human preference: 64.2

Not drawn, measured for fewer than two of these models: Agents & tools, Knowledge, Instruction following, Long context.

ModelIndex

Where sources disagree

Across the 13 benchmarks both have, Claude Opus 4.8 leads 8, GPT-6 Sol leads 5. Every overall index that rates both puts Claude Opus 4.8 ahead.

Overall indexClaude Opus 4.8 (high)GPT-6 Sol (max)Leader
BenchLeader Index60.562.9neck and neck
Arena Text14821457Claude Opus 4.8

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 4.8 (high)GPT-6 Sol (max)
Blended price per 1M tokens$10.00$4.00
Input / output per 1M$5.00 / $25.00$2.00 / $10.00
Cached input per 1M$0.500$0.200
Price per task: chat reply$0.0095$0.0038
Price per task: summarize a 30-page report$0.075$0.030
Price per task: code edit$0.068$0.027
Price per task: agentic coding session$0.400$0.160
Price per task: structured extraction$0.015$0.0060
Agentic coding session, input mostly cached$0.198$0.079
Cost per benchmark task (AA Index)–$2.47
Output tokens per task (AA Index)–54k
Of which reasoning–35k
Time per task (AA Index)–7 min
Output speed71 tok/s121 tok/s
Time to first answer0.8 s72.1 s
Time to a complete answer–130.2 s

Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 4.8 (high) scores 60.5 and takes 1 s to its first answer token; Claude Opus 4.8 (max) scores 60.2 in 37 s, Claude Opus 4.8 scores 60.1 in 1 s. GPT-6 Sol (max) scores 62.9 and takes 72 s to its first answer token; GPT-6 Sol (xhigh) scores 58.6 in 24 s, GPT-6 Sol (high) scores 58.2 in 9 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

145 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 4.8 (high)GPT-6 Sol (max)
BenchLeader IndexBenchLeader60.562.9
Coding scoreBenchLeader61.466.5
Human preference scoreBenchLeader67.364.2
Maths scoreBenchLeader67.468.5
Multimodal scoreBenchLeader64.060.7
Reasoning scoreBenchLeader60.066.2
Agents & tools scoreBenchLeader–59.6
Knowledge scoreBenchLeader–61.5
Long context scoreBenchLeader–66.6
Blended price $/MBenchLeader$10.00$4.00
Output speedBenchLeader71 tok/s121 tok/s
Time to first answerBenchLeader0.8 s72.1 s
Context windowBenchLeader1,000,000 tokens1,050,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub91.0%†94.3%
SimpleBenchSimpleBench64.8%†–
Arena Hard PromptsArena (LMArena)15131483
LiveBench 1.0 ReasoningLiveBench89.2%†88.7%
GPQA Diamond (AA)Artificial Analysis92.0%†–
Humanity's Last Exam (AA)Artificial Analysis48.7%†47.9%
HLE-Diamond (AA)Artificial Analysis–45.2%
GPQA Diamond (Vals)Vals AI92.4%†–
EnigmaEvalScale AI SEAL23.5%†–
Kagi LLM BenchmarkKagi LLM Benchmark88.8%†–
ARC-AGI-1ARC Prize92.0%95.5%
ARC-AGI-2ARC Prize72.1%89.6%
ARC-AGI-3ARC Prize1.5%23.0%
CritPtArtificial Analysis20.9%†30.9%
CritPt v2 hard subset (AA)Artificial Analysis–66.5%
MysteryMechanismVals AI–30.2%
Chess PuzzlesEpoch AI Benchmarking Hub52.0%†56.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub36.0%†56.0%
LMCAEpoch AI Benchmarking Hub57.5%†59.1%
DTBenchEpoch AI Benchmarking Hub94.9%†97.3%
ForecastBenchForecastBench62.7%†–
BTF-3Epoch AI Benchmarking Hub14.0%–
Coding
SciCodeSciCode53.5%†57.6%
WeirdMLWeirdML70.5%†–
FrontierCodeCognition46.5%†49.3%
Show all 145 benchmarks
MetricPublisherClaude Opus 4.8 (high)GPT-6 Sol (max)
Coding
GSO-BenchGSO-Bench47.1%†–
Arena CodingArena (LMArena)15341520
Arena WebDevArena (LMArena)15551674
LiveBench 1.0 CodingLiveBench81.8%†81.8%
SciCode (AA)Artificial Analysis54.4%†57.6%
LiveCodeBenchVals AI87.8%†–
IOIVals AI–82.6%
SWE-bench (Vals)Vals AI88.6%†–
Code MigrationVals AI47.3%†57.2%
ProgramBenchVals AI1.0%†2.0%
Vibe Code Bench v1.1Vals AI82.7%†87.8%
SWE Atlas: Codebase QnAScale AI SEAL57.3%†–
SWE Atlas: RefactoringScale AI SEAL46.7%†–
SWE Atlas: Test WritingScale AI SEAL49.6%†–
Surface Evolver BenchEpoch AI Benchmarking Hub87.5%–
DeepSWE v1.1Epoch AI Benchmarking Hub51.8%–
ALE-BenchEpoch AI Benchmarking Hub1563.82462
Terminal-Bench 4.0 (AA)Artificial Analysis21.7%†43.9%
Terminal-Bench 2.1 (AA)Artificial Analysis84.6%†–
GameBench 2 OverallDuelLab GameBench 249.6†72.9†
GameBench 2 by reasoning settingDuelLab GameBench 245.4†79.3†
Design Arena WebsiteDesign Arena1256†1285†
FrontierCode 2.0 (AA)Artificial Analysis–35.6%
Agents & tools
Terminal-BenchTerminal-Bench23.6%†50.3%
OSWorld-Verified 2.0OSWorld20.6%†–
Remote Labor IndexScale AI / CAIS8.3%†–
APEX-AgentsMercor48.9%†54.3%
Arena AgentArena (LMArena)–7
LiveBench Agentic CodingLiveBench50.5%†52.9%
Terminal-Bench HardArtificial Analysis58.3%†–
τ²-Bench Telecom (AA)Artificial Analysis94.4%†–
Terminal-Bench 2.1 (Vals)Vals AI71.9%†83.2%
MCP AtlasScale AI SEAL82.2%†–
HiL-BenchScale AI SEAL35.3%†–
τ³-bench Bankingτ²-bench39.7%†–
τ³-Banking (AA)Artificial Analysis34.2%†–
ITBench SRE (AA)Artificial Analysis–49.4%
Analyst Agent (AA)Artificial Analysis45.0%†–
CyberBenchVals AI–78.0%
Finance Agent v2Vals AI53.9%†49.0%
Harvey's Legal Agent Benchmark v1Vals AI9.6%†1.7%
Legal Research BenchVals AI43.8%†28.9%
SkillsBenchVals AI59.2%†–
SREBenchVals AI–30.5%
Tax Agent BenchVals AI30.0%†15.0%
Terminal-Bench 2.0 (Vals)Vals AI70.0%†–
Terminal-Bench 4.0 (Vals)Vals AI23.2%†44.4%
Terminal-Bench ScienceVals AI4.3%†30.0%
Time Horizon Index: KSPVals AI11.8%†–
Arena SearchArena (LMArena)1194†1229
DeepResearch BenchDeepResearch Bench50.2%–
PostTrainBenchEpoch AI Benchmarking Hub33.8%–
GBAEvalEpoch AI Benchmarking Hub70.9%†–
Vending-Bench 2Andon Labs5787.4†14427.9†
AutomationBenchArtificial Analysis45.6%†61.6%
GDP.pdf (AA)Artificial Analysis22.8%†25.2%
Harvey LAB-AA v1.1Harvey–3.6%
EnterpriseOps-GymArtificial Analysis44.0%†–
AA-Briefcase v1.1Artificial Analysis1320†1476
Design Arena AgenticDesign Arena1221†1212†
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–30.0%
CWE-Bench (AA)Artificial Analysis–64.2%
DeepSec-Bench (AA)Artificial Analysis–46.3%
CyberGym end-to-end (AA)Artificial Analysis–0.0%
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub80.0%†89.8%
FrontierMath Tier 4Epoch AI Benchmarking Hub56.1%†90.0%
OTIS Mock AIMEEpoch AI Benchmarking Hub98.3%†100.0%
ProofBenchVals AI69.0%†83.0%†
LiveBench MathematicsLiveBench94.3%†96.4%
AIME 2026MathArena100.0%†–
HMMT February 2026MathArena95.5%†–
MathArena ApexMathArena81.3%†–
Arena MathsArena (LMArena)14921440
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub53.0%†60.7%
LiveBench 1.0 Data AnalysisLiveBench66.0%†81.2%
AA-Omniscience v1.1Artificial Analysis–21.4
MMLU-ProVals AI89.6%†–
LegalBenchVals AI83.6%†–
CorpFinVals AI66.7%†–
TaxEvalVals AI75.6%†–
PRBench LegalScale AI SEAL–38.6%†
BioMysteryBenchVals AI–74.8%
Excel Modeling BenchmarkVals AI69.4%†71.5%
MedCodeVals AI53.2%†47.1%
MedScribeVals AI85.8%†82.0%
Public Benefits Bench v1.1Vals AI68.1%†56.6%
DrugDiscoveryBenchScale AI SEAL46.8%†–
EBR-benchEpoch AI Benchmarking Hub28.6%†53.3%
GDP.pdfEpoch AI Benchmarking Hub24.0%†26.4%
AA-Omniscience v1.1: accuracyArtificial Analysis–51.1%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis–39.1%
HealthBench Professional (AA)OpenAI–62.3%
Instruction following
LiveBench LanguageLiveBench79.7%†85.3%
LiveBench Instruction FollowingLiveBench72.0%†68.6%
IFBenchArtificial Analysis62.2%†–
Arena Instruction FollowingArena (LMArena)14921448
Human preference
Arena TextArena (LMArena)14821457
EQ-Bench 4EQ-Bench1268†–
Arena Creative WritingArena (LMArena)14701430
Arena Multi-turnArena (LMArena)14971466
Arena Longer QueriesArena (LMArena)15031466
Multimodal
Arena VisionArena (LMArena)12951232
MMMU-ProArtificial Analysis–83.0%
MMMU-Pro (Vals)Vals AI86.6%†–
MortgageTaxVals AI69.9%†–
Blueprint-Bench 2Epoch AI Benchmarking Hub14.5%†36.9%†
AA-3DArtificial Analysis–1131
Long context
AA-LCR v1.1Artificial Analysis77.7%†83.7%
Arena DocumentArena (LMArena)1485–
MLCRArtificial Analysis19.4%†33.9%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub159.9†164.1†
LiveBench 1.0LiveBench76.2%†79.3%
AA Intelligence Index v5.0Artificial Analysis–44.2
AA Intelligence Index v4.3.2Artificial Analysis41.8†47.6
AA Progress Index v1Artificial Analysis318.4†334
Vals RSI IndexVals AI24.0%†28.1%
Vals Science IndexVals AI–45.0%
Vals Multimodal IndexVals AI70.9%†–
AA Cyber IndexArtificial Analysis–36.8
Safety & honesty
FORTRESSScale AI SEAL18.2%†13.5%†

Data as of 2026-11-09. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261103-eaea31df94e6 (https://benchmarks.duellab.org/releases/gb2-20261103-eaea31df94e6/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • GPT-6 Sol (max): BenchLeader Index 63.5 → 62.9.
  • GPT-6 Sol (max): 2 score revisions.
  • GPT-6 Sol (max): 1 score revision.
  • GPT-6 Sol (max): 1 score revision.
  • GPT-6 Sol (max): 6 new results (AA Intelligence Index v5.0, AA-Omniscience v1.1, CritPt v2 hard subset (AA)…).
  • GPT-6 Sol (max): 1 score revision.
  • GPT-6 Sol (max): 1 score revision.
  • GPT-6 Sol (max): BenchLeader Index 64.1 → 63.5.
  • GPT-6 Sol (max): 2 new results (AA Progress Index v1, HLE-Diamond (AA)).
  • GPT-6 Sol (max): 1 score revision.

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

Claude Opus 4.8 vs GPT-6 Sol: questions

Is Claude Opus 4.8 better than GPT-6 Sol?
Claude Opus 4.8 (high) and GPT-6 Sol (max) are neck and neck on quality (60.5 vs 62.9, within the ±6.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 9 Nov 2026, so check the category scores for your use.
Claude Opus 4.8 vs GPT-6 Sol: which is cheaper, and will that change?
GPT-6 Sol is cheaper today: $4.00 against $10.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Claude Opus 4.8 or GPT-6 Sol better for coding?
GPT-6 Sol scores higher in coding (67 vs 61 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1555 for Claude Opus 4.8 against 1674 for GPT-6 Sol.
Which is faster, Claude Opus 4.8 or GPT-6 Sol?
GPT-6 Sol streams ~1.5× faster (121 vs 71 output tokens per second, Artificial Analysis 7-day median).
Which is better for long documents?
Both take about 1.1M tokens: Claude Opus 4.8 1,000,000, GPT-6 Sol 1,050,000.
More questions (3)
Which has lower latency?
Claude Opus 4.8 answers first: 0.8 s against 72.1 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 4.8 vs GPT-6 Sol?
Across the 13 benchmarks both have, Claude Opus 4.8 leads 8, GPT-6 Sol leads 5. Every overall index that rates both puts Claude Opus 4.8 ahead.
When was GPT-6 Sol released?
GPT-6 Sol was released on 22 Sep 2026.