BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases12

Claude Opus 5.5 vs Grok 4.6

Full profiles: Claude Opus 5.5 (Anthropic) and Grok 4.6 (SpaceXAI): every benchmark, price and rank.

Updated · Data from 12 sources

As of 9 Nov 2026, Claude Opus 5.5 is ahead of Grok 4.6 on quality (68.9 vs 60.5 on the BenchLeader Index), but Grok 4.6 is 2.7× cheaper at $3.00 vs $8.00 per 1M tokens blended. Claude Opus 5.5 leads in every category we measure both in. Claude Opus 5.5 streams faster.

Key facts (as of 9 Nov 2026)

  • Claude Opus 5.5 scores 68.9 on the BenchLeader Index and Grok 4.6 scores 60.5 (BenchLeader, 9 Nov 2026).
  • Claude Opus 5.5 costs $8.00 per 1M tokens blended (list price, 9 Nov 2026).
  • Grok 4.6 costs $3.00 per 1M tokens blended (list price, 9 Nov 2026).
  • Claude Opus 5.5 streams faster than Grok 4.6 (Artificial Analysis, 7-day median, 9 Nov 2026).
  • Of the 22 benchmarks both models have, Claude Opus 5.5 leads 21 (BenchLeader, 9 Nov 2026).
  • On Arena WebDev, Claude Opus 5.5 is rated 1818 and Grok 4.6 1618 (LMArena, as published by 9 Nov 2026).
Claude Opus 5.5 benchmarks max Anthropic Grok 4.6 benchmarks high SpaceXAIAdd 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
CodingClaude Opus 5.580 vs 61 category score; Arena WebDev 1818 vs 1618High
Agents & tool useClaude Opus 5.563 vs 58 category score; Terminal-Bench 65.8% vs 19.7%Medium
MathsClaude Opus 5.574 vs 63 category score; OTIS Mock AIME 100.0% vs 97.8%High
ReasoningClaude Opus 5.572 vs 64 category score; ARC-AGI-2 91.7% vs 65.1%High
KnowledgeClaude Opus 5.574 vs 60 category score; SimpleQA Verified 72.2% vs 49.3%High
Long documentsClaude Opus 5.567 vs 65 category score; AA-LCR v1.1 84.7% vs 80.3%Medium
Images and other mediaClaude Opus 5.570 vs 60 category scoreHigh
Chat quality (human preference)–Only Grok 4.6 has an Arena rating (64 category score)Not comparable
Lowest costGrok 4.6$3.00 vs $8.00 per 1M blended; agentic coding session $0.144 vs $0.320–
Fastest first answerGrok 4.629.1 s (Grok 4.6 (high)) vs 380.6 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: 68.9 vs 60.5.
  • Claude Opus 5.5 (max) is stronger in agents & tools, coding, knowledge, long context, maths, multimodal, reasoning.
  • Grok 4.6 (high) is stronger in human preference.
  • Grok 4.6 (high) is 2.7× cheaper ($3.00 vs $8.00 per 1M blended).
  • Claude Opus 5.5 (max) streams 1.6× faster (115 vs 72 tokens per second).

At a glance

SpecClaude Opus 5.5Grok 4.6
DeveloperAnthropicSpaceXAI
Released22 Sep 202611 Aug 2026
Context window (tokens)1,000,000500,000
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)Grok 4.6 (high)
Open weightsNoNo
Price in / out / cached input per 1M$4.00 / $20.00 / $0.200$2.00 / $6.00 / $0.500
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)Grok 4.6 (high)Leader
Reasoning72.363.6Claude Opus 5.5
Coding80.461.4Claude Opus 5.5
Agents & tools62.757.9Claude Opus 5.5
Maths74.063.4Claude Opus 5.5
Knowledge73.759.8Claude Opus 5.5
Human preference–64.2not comparable
Multimodal70.360.0Claude Opus 5.5
Long context67.164.9Claude Opus 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 & toolsMathsKnowledgeLong contextMultimodalClaude Opus 5.5, Reasoning: 72.3Claude Opus 5.5, Coding: 80.4Claude Opus 5.5, Agents & tools: 62.7Claude Opus 5.5, Maths: 74.0Claude Opus 5.5, Knowledge: 73.7Claude Opus 5.5, Long context: 67.1Claude Opus 5.5, Multimodal: 70.3Grok 4.6, Reasoning: 63.6Grok 4.6, Coding: 61.4Grok 4.6, Agents & tools: 57.9Grok 4.6, Maths: 63.4Grok 4.6, Knowledge: 59.8Grok 4.6, Long context: 64.9Grok 4.6, Multimodal: 60.0

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

ModelIndex

Where sources disagree

Across the 22 benchmarks both have, Claude Opus 5.5 leads 21, Grok 4.6 leads 1. Every overall index that rates both puts Claude Opus 5.5 ahead.

Overall indexClaude Opus 5.5 (max)Grok 4.6 (high)Leader
BenchLeader Index68.960.5Claude Opus 5.5
AA Intelligence Index v5.059.131.0†Claude Opus 5.5
Arena Text1506†1456Claude 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)Grok 4.6 (high)
Blended price per 1M tokens$8.00$3.00
Input / output per 1M$4.00 / $20.00$2.00 / $6.00
Cached input per 1M$0.200$0.500
Price per task: chat reply$0.0076$0.0026
Price per task: summarize a 30-page report$0.060$0.028
Price per task: code edit$0.054$0.021
Price per task: agentic coding session$0.320$0.144
Price per task: structured extraction$0.012$0.0052
Agentic coding session, input mostly cached$0.149$0.076
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/s72 tok/s
Time to first answer380.6 s29.1 s
Time to a complete answer771.1 s57.7 s

Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 5.5 (max) scores 68.9 and takes 381 s to its first answer token; Claude Opus 5.5 (xhigh) scores 65.5 in 84 s, Claude Opus 5.5 (high) scores 65.4 in 28 s. Grok 4.6 (high) scores 60.5 and takes 29 s to its first answer token; Grok 4.6 (medium) scores 59.5 in 22 s, Grok 4.6 (xhigh) scores 57.6 in 29 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

152 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)Grok 4.6 (high)
BenchLeader IndexBenchLeader68.960.5
Agents & tools scoreBenchLeader62.757.9
Coding scoreBenchLeader80.461.4
Knowledge scoreBenchLeader73.759.8
Long context scoreBenchLeader67.164.9
Maths scoreBenchLeader74.063.4
Multimodal scoreBenchLeader70.360.0
Reasoning scoreBenchLeader72.363.6
Human preference scoreBenchLeader–64.2
Blended price $/MBenchLeader$8.00$3.00
Output speedBenchLeader115 tok/s72 tok/s
Time to first answerBenchLeader380.6 s29.1 s
Context windowBenchLeader1,000,000 tokens500,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub90.6%94.0%
SimpleBenchSimpleBench–75.9%†
Arena Hard PromptsArena (LMArena)1530†1481
LiveBench 2.0 ReasoningLiveBench92.2%†91.2%†
LiveBench 1.0 ReasoningLiveBench92.2%90.5%†
GPQA Diamond (AA)Artificial Analysis–95.0%
Humanity's Last Exam (AA)Artificial Analysis61.4%42.9%
HLE-Diamond (AA)Artificial Analysis67.9%21.5%†
GPQA Diamond (Vals)Vals AI–94.7%
ARC-AGI-1ARC Prize97.5%87.0%
ARC-AGI-2ARC Prize91.7%65.1%
ARC-AGI-3ARC Prize–2.1%†
CritPtArtificial Analysis31.7%17.1%
CritPt v2 hard subset (AA)Artificial Analysis68.0%40.0%†
MysteryMechanismVals AI49.5%†30.6%
Chess PuzzlesEpoch AI Benchmarking Hub67.0%33.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub71.0%34.0%†
LMCAEpoch AI Benchmarking Hub68.2%48.5%†
DTBenchEpoch AI Benchmarking Hub98.9%97.3%†
ForecastBenchForecastBench–65.5%†
Coding
SciCodeSciCode66.9%56.5%
WeirdMLWeirdML–67.3%
FrontierCodeCognition54.6%†48.0%†
Arena CodingArena (LMArena)1546†1508
Arena WebDevArena (LMArena)18181618
Show all 152 benchmarks
MetricPublisherClaude Opus 5.5 (max)Grok 4.6 (high)
Coding
LiveBench 2.0 CodingLiveBench87.1%†76.7%†
LiveBench 1.0 CodingLiveBench89.3%76.8%†
LiveCodeBenchVals AI–88.2%
IOIVals AI95.1%†47.6%
SWE-bench (Vals)Vals AI–95.6%
Code MigrationVals AI66.7%†44.6%
Code Migration v1.1Vals AI81.0%†–
ProgramBenchVals AI18.5%†1.0%
SWE-bench MultimodalVals AI70.2%†–
Vibe Code Bench 1-100Vals AI30.4%†14.8%
Vibe Code Bench v1.1Vals AI90.3%†76.2%
MirrorCodeEpoch AI Benchmarking Hub78.9%†41.1%†
DeepSWE v1.1Epoch AI Benchmarking Hub–65.2%
CursorBenchCursor57.8%40.4%
ALE-BenchEpoch AI Benchmarking Hub2147.4†1508.1†
FrontierSWEEpoch AI Benchmarking Hub62.3%25.3%†
Terminal-Bench 4.0 (AA)Artificial Analysis59.6%21.2%
Terminal-Bench 2.1 (AA)Artificial Analysis–88.4%
Startrise FrontendStartrise71.1%†72.1%†
GameBench 2 OverallDuelLab GameBench 287.6†59.0†
GameBench 2 by reasoning settingDuelLab GameBench 294.8†68.6†
Design Arena WebsiteDesign Arena1325†1284†
FrontierCode 1.1 (AA)Artificial Analysis52.2%–
FrontierCode 2.0 (AA)Artificial Analysis46.4%19.8%†
Agents & tools
Terminal-BenchTerminal-Bench65.8%19.7%
Remote Labor IndexScale AI / CAIS21.3%†–
APEX-AgentsMercor73.5%65.3%†
Arena AgentArena (LMArena)20-0.2†
LiveBench Long AgenticLiveBench72.8%†18.7%†
LiveBench BrowserLiveBench88.4%†49.4%†
LiveBench Agentic CodingLiveBench71.7%57.0%†
Terminal-Bench 2.1 (Vals)Vals AI87.6%†78.3%
MCP AtlasScale AI SEAL85.4%†–
HiL-BenchScale AI SEAL59.0%†–
GDPval-AA v2.1, normalised scoreArtificial Analysis68.3%–
GDPval-AA v2.1Artificial Analysis1866–
τ³-Banking (AA)Artificial Analysis–50.7%
ITBench SRE (AA)Artificial Analysis38.2%–
Analyst Agent (AA)Artificial Analysis56.3%41.3%
CUA-benchVals AI14.0%–
CyberBenchVals AI55.4%†67.7%
Finance Agent v2Vals AI58.6%†53.7%
Finance Agent v2.1Vals AI67.8%†–
Harvey's Legal Agent Benchmark v1.1Vals AI9.2%†–
Harvey's Legal Agent Benchmark v1Vals AI3.8%†15.8%
Legal Research BenchVals AI50.5%†48.1%
Robo-benchVals AI14.6%–
SkillsBenchVals AI–55.8%
SREBenchVals AI33.6%†–
Tax Agent BenchVals AI44.5%†33.7%
Terminal-Bench 4.0 (Vals)Vals AI65.2%†17.2%
Terminal-Bench ScienceVals AI47.1%†11.4%
Time Horizon Index: KSPVals AI91.3%†5.8%
Arena SearchArena (LMArena)1214†–
Vending-Bench 2Andon Labs9235.3†9047.0†
AutomationBenchArtificial Analysis69.5%66.7%
GDP.pdf (AA)Artificial Analysis26.2%17.0%
Harvey LAB-AA v1.1Harvey4.2%–
EnterpriseOps-GymArtificial Analysis–48.3%
AA-Briefcase v1.1Artificial Analysis18001528
GDPxGDPx by SemiAnalysis12.0%–
Design Arena AgenticDesign Arena1334†1233†
Terminal-Bench-Science 0.1 (AA)Artificial Analysis59.0%6.2%
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%66.0%†
FrontierMath Tier 4Epoch AI Benchmarking Hub95.0%31.7%†
OTIS Mock AIMEEpoch AI Benchmarking Hub100.0%97.8%
ProofBenchVals AI100.0%51.0%†
LiveBench MathematicsLiveBench97.1%92.6%†
Arena MathsArena (LMArena)1518†1444
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub72.2%49.3%
LiveBench 2.0 Data AnalysisLiveBench80.3%†75.3%†
LiveBench ResearchLiveBench88.9%†88.8%†
LiveBench 1.0 Data AnalysisLiveBench80.3%73.9%†
AA-Omniscience v1.1Artificial Analysis48.124.9†
MMLU-ProVals AI–89.4%
LegalBenchVals AI–86.3%
CorpFinVals AI–66.2%
TaxEvalVals AI–71.1%
PRBench LegalScale AI SEAL47.6%†–
BioMysteryBenchVals AI79.3%†72.2%
ClaimBenchVals AI100.0%†–
Excel Modeling BenchmarkVals AI75.9%†62.7%
Excel Modeling Benchmark v1.1Vals AI75.8%†–
MedCodeVals AI49.8%†44.7%
MedScribeVals AI91.4%†86.5%
Public Benefits Bench v1.1Vals AI70.6%†66.8%
EBR-benchEpoch AI Benchmarking Hub71.4%30.5%†
GDP.pdfEpoch AI Benchmarking Hub30.6%16.0%
AA-Omniscience v1.1: accuracyArtificial Analysis65.5%46.2%†
AA-Omniscience v1.1: non-hallucinationArtificial Analysis49.7%60.3%†
HealthBench Professional (AA)OpenAI59.9%–
Instruction following
LiveBench LanguageLiveBench86.3%83.7%†
LiveBench Instruction FollowingLiveBench65.7%71.9%†
Arena Instruction FollowingArena (LMArena)1519†1451
Human preference
Arena TextArena (LMArena)1506†1456
EQ-Bench 4EQ-Bench1340†–
Arena Creative WritingArena (LMArena)1517†1443
Arena Multi-turnArena (LMArena)1500†1449
Arena Longer QueriesArena (LMArena)1520†1476
Multimodal
Arena VisionArena (LMArena)1324†1264
LiveBench DesignLiveBench83.3%†76.3%†
MMMU-ProArtificial Analysis87.7%–
MortgageTaxVals AI–64.2%
SAGEVals AI91.3%†–
Blueprint-Bench 2Epoch AI Benchmarking Hub51.2%†33.2%†
Code2Video-AAArtificial Analysis1536–
AA-3DArtificial Analysis1202–
Long context
AA-LCR v1.1Artificial Analysis84.7%80.3%
Arena DocumentArena (LMArena)–1451
MLCRArtificial Analysis27.8%15.6%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub168.4†157.9†
LiveBench 2.0LiveBench83.4%†63.5%†
LiveBench 1.0LiveBench83.2%78.0%†
AA Intelligence Index v5.0Artificial Analysis59.131.0†
AA Intelligence Index v4.3.2Artificial Analysis57.644.3
AA Progress Index v1Artificial Analysis350321.5
Vals IndexVals AI70.1†–
Vals Finance IndexVals AI65.3%†–
Vals RSI IndexVals AI37.3%25.5%†
Vals Science IndexVals AI58.6%†38.1%
Vals Router IndexVals AI64.2%†–
AA Cyber IndexArtificial Analysis28.7–
Safety & honesty
FORTRESSScale AI SEAL5.8%–

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

  • Claude Opus 5.5 (max): 2 score revisions.
  • Grok 4.6 (high): 2 score revisions.
  • Claude Opus 5.5 (max): 1 score revision.
  • Grok 4.6 (high): 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.
  • Claude Opus 5.5 (max): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 3 score revisions; cost per task moved from $5.98/M to $10.04/M (68% up).
  • Grok 4.6 (high): 1 new result (AA Progress Index v1).

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

Claude Opus 5.5 vs Grok 4.6: questions

Is Claude Opus 5.5 better than Grok 4.6?
Claude Opus 5.5 (max) leads on quality: 68.9 vs 60.5. The BenchLeader Index combines every independent quality benchmark; Claude Opus 5.5 (max) is ahead overall as of 9 Nov 2026, but check the category scores for your use.
Claude Opus 5.5 vs Grok 4.6: which is cheaper, and will that change?
Grok 4.6 is cheaper today: $3.00 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 Grok 4.6 better for coding?
Claude Opus 5.5 scores higher in coding (80 vs 61 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1818 for Claude Opus 5.5 against 1618 for Grok 4.6.
Which is faster, Claude Opus 5.5 or Grok 4.6?
Claude Opus 5.5 streams faster (115 vs 72 output tokens per second, Artificial Analysis 7-day median).
Is Claude Opus 5.5 or Grok 4.6 better for agents and tool use?
Claude Opus 5.5 scores higher in agentic tasks (63 vs 58 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 65.8% for Claude Opus 5.5 against 19.7% for Grok 4.6.
More questions (4)
Which is better for long documents?
Claude Opus 5.5 scores higher on long-context benchmarks (67 vs 65). Claude Opus 5.5 accepts more context: 1,000,000 against 500,000 tokens for Grok 4.6.
Which has lower latency?
Grok 4.6 answers first: 29.1 s against 380.6 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 Grok 4.6?
Across the 22 benchmarks both have, Claude Opus 5.5 leads 21, Grok 4.6 leads 1. 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.