VerdictPal
Join

Models / Anthropic

Claude Opus 4.6

Claude Opus · Released 2026-02-05

Anthropic's February 2026 flagship. First Opus with 1M context (beta), 128K output, multi-agent teams in Claude Code. $5/$25 per 1M tokens.

Frontier Opus model for complex reasoning, multi-step coding, and massive context tasks. Superseded by Opus 4.7 but still available.

#67 of 168 on AA Index · current snapshot
39 AA Index · 16 benchmark rows · 7 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite56
Context1M
Input / 1M$5.0
Output / 1M$25
Knowledge cutoff2025-12-01
Statussolid

Family profile

Best published score in each covered benchmark family.

76/100 avg
Knowledge
91
Reasoning
84
Math
41
Coding
96
Agentic
80
Long context
62

6 tested benchmark families

Benchmark placements

Where Claude Opus 4.6 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
HumanEvalCoding#1/ 1596Anthropic Claude 3.7/4.6 model card2026-01-01
LiveBenchReasoning#1/ 2171LiveBench leaderboard2026-04-15
MMLUKnowledge#2/ 2291Anthropic Claude 3.7/4.6 model card2026-01-01
Vending-Bench 2Agentic#5/ 10Andon Labs Vending-Bench 22026-08-04
SWE-bench VerifiedCoding#6/ 2381SWE-bench official leaderboard2026-05-30
FrontierMathMath#7/ 1541Epoch AI FrontierMath Tiers 1-3 (v2) CSV2026-08-15
ARC-AGIReasoning#13/ 141ARC-AGI-3 paper (arXiv 2603.24621)2026-03-20
Terminal-BenchAgentic#25/ 18480Terminal-Bench 2.0 leaderboard2026-05-20
SciCodeCoding#75/ 18046Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#83/ 18739Artificial Analysis2026-09-01
AA time to first tokenPerformance#94/ 186Artificial Analysis2026-09-01
AA output speedPerformance#95/ 187Artificial Analysis2026-09-01
GPQA DiamondReasoning#109/ 18584Artificial Analysis2026-09-01
IFBenchReasoning#110/ 12745Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#125/ 18019Artificial Analysis2026-09-01
AA-LCRLong context#131/ 17962Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where Claude Opus 4.6 shows up, it's highlighted.

KnowledgeClaude Opus 4.6 · 91
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningClaude Opus 4.6 · 84
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
MathClaude Opus 4.6 · 41
  1. 1
    OpenAI o3
    99
  2. 2
    GPT-5 (medium)
    99
  3. 3
    GPT-5 (high)
    99
CodingClaude Opus 4.6 · 96
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticClaude Opus 4.6 · 80
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextClaude Opus 4.6 · 62
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for Claude Opus 4.6 in the last 120 days. Older rows live in the full table below.

12 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context62.33%Artificial Analysis5
AA output speedPerformance0 t/sArtificial Analysis5
AA time to first tokenPerformance0.00sArtificial Analysis5
Artificial Analysis Intelligence IndexReasoning38.8%Artificial Analysis5
GPQA DiamondReasoning84%Artificial Analysis5
Humanity's Last ExamReasoning19.1%Artificial Analysis5
IFBenchReasoning44.63%Artificial Analysis5
SciCodeCoding45.7%Artificial Analysis5
FrontierMathMath40.7%Epoch AI FrontierMath Tiers 1-3 (v2) CSV22
Vending-Bench 2Agentic8017.59Andon Labs Vending-Bench 233
SWE-bench VerifiedCoding80.8%SWE-bench official leaderboard99
Terminal-BenchAgentic79.8%Terminal-Bench 2.0 leaderboard109
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Knowledge91
Reasoning84
Math41
Coding96
Agentic80
Long context62
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Claude Opus 4.6. Scores link to the original source; gaps mean no public row exists yet.

16 sourced rows

16 of 37 catalog benchmarks have a sourced row for Claude Opus 4.6.

Knowledge

Reasoning

Math

Coding

Agentic

Long context

Performance

Changelog
  1. Released Claude Opus 4.6 — first Opus with 1M context, 128K output, agent teams.
Sources