VerdictPal
Join

Models / Anthropic

Claude Opus 4.5

Claude Opus · Released 2025-11-24

Anthropic's November 2025 flagship. Best at coding and agents at its release. $5/$25 per 1M tokens, 200K context. Superseded by Opus 4.6/4.7.

Previous-gen Opus for coding and agentic tasks. Still available but Opus 4.7 is the current default.

#84 of 168 on AA Index · current snapshot
36 AA Index · 13 benchmark rows · 2 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite56
Context200K
Input / 1M$5.0
Output / 1M$25
Knowledge cutoff2025-10-01
Statussolid

Family profile

Best published score in each covered benchmark family.

70/100 avg
Knowledge
89
Reasoning
81
Math
63
Coding
81
Agentic
41
Long context
67

6 tested benchmark families

Benchmark placements

Where Claude Opus 4.5 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
MMLU-ProKnowledge#5/ 3989Artificial Analysis2026-09-01
SWE-bench VerifiedCoding#5/ 2381SWE-bench SEAL leaderboard2026-05-30
LiveCodeBenchCoding#21/ 3174Artificial Analysis2026-09-01
AIMEMath#31/ 3763Artificial Analysis2026-09-01
SciCodeCoding#62/ 18047Artificial Analysis2026-09-01
AA time to first tokenPerformance#92/ 186Artificial Analysis2026-09-01
AA output speedPerformance#93/ 187Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#105/ 18736Artificial Analysis2026-09-01
AA-LCRLong context#109/ 17967Artificial Analysis2026-09-01
Terminal-BenchAgentic#111/ 18441Artificial Analysis2026-09-01
IFBenchReasoning#115/ 12743Artificial Analysis2026-09-01
GPQA DiamondReasoning#133/ 18581Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#145/ 18013Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where Claude Opus 4.5 shows up, it's highlighted.

KnowledgeClaude Opus 4.5 · 89
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningClaude Opus 4.5 · 81
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
MathClaude Opus 4.5 · 63
  1. 1
    OpenAI o3
    99
  2. 2
    GPT-5 (medium)
    99
  3. 3
    GPT-5 (high)
    99
CodingClaude Opus 4.5 · 81
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticClaude Opus 4.5 · 41
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextClaude Opus 4.5 · 67
  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 4.5 in the last 120 days. Older rows live in the full table below.

13 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context67.33%Artificial Analysis6
AA output speedPerformance0 t/sArtificial Analysis6
AA time to first tokenPerformance0.00sArtificial Analysis6
AIMEMath62.67%Artificial Analysis6
Artificial Analysis Intelligence IndexReasoning35.6%Artificial Analysis6
GPQA DiamondReasoning81%Artificial Analysis6
Humanity's Last ExamReasoning13.2%Artificial Analysis6
IFBenchReasoning42.99%Artificial Analysis6
LiveCodeBenchCoding73.8%Artificial Analysis6
MMLU-ProKnowledge88.9%Artificial Analysis6
SciCodeCoding47%Artificial Analysis6
Terminal-BenchAgentic40.91%Artificial Analysis6
SWE-bench VerifiedCoding80.9%SWE-bench SEAL leaderboard100
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.

Knowledge89
Reasoning81
Math63
Coding81
Agentic41
Long context67
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 4.5. Scores link to the original source; gaps mean no public row exists yet.

13 sourced rows

13 of 37 catalog benchmarks have a sourced row for Claude Opus 4.5.

Knowledge

Reasoning

Math

Coding

Agentic

Long context

Performance

Changelog
  1. Released Claude Opus 4.5 — best coding and agent model at launch, $5/$25.
Sources