VerdictPal
Join

Models / Anthropic

Claude 4.5 Sonnet (Reasoning)

Claude 4.5 · Released 2025-09-29

AA Intelligence Index 36.4. Listed on Artificial Analysis with API pricing $3/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#130 of 184 on AA Index · current snapshot
21 AA Index · 13 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite55
Context128K
Input / 1M$3.0
Output / 1M$15
Knowledge cutoffopen
Statussolid

Family profile

Best published score in each covered benchmark family.

76/100 avg
Knowledge
88
Reasoning
83
Math
88
Coding
71
Agentic
56
Long context
72

6 tested benchmark families

Benchmark placements

Where Claude 4.5 Sonnet (Reasoning) 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#8/ 3988Artificial Analysis2026-09-10
AIMEMath#21/ 3788Artificial Analysis2026-09-10
LiveCodeBenchCoding#22/ 3171Artificial Analysis2026-09-10
τ³-Bench BankingAgentic#52/ 11125Artificial Analysis2026-09-10
SciCodeCoding#56/ 9446Artificial Analysis2026-09-10
Terminal-BenchAgentic#80/ 18956Artificial Analysis2026-09-10
IFBenchReasoning#87/ 12757Artificial Analysis2026-09-10
AA output speedPerformance#91/ 190Artificial Analysis2026-09-10
AA time to first tokenPerformance#91/ 190Artificial Analysis2026-09-10
AA-LCRLong context#104/ 18272Artificial Analysis2026-09-10
GPQA DiamondReasoning#121/ 19083Artificial Analysis2026-09-10
Artificial Analysis Intelligence IndexReasoning#132/ 18821Artificial Analysis2026-09-10
Humanity's Last ExamReasoning#134/ 18518Artificial Analysis2026-09-10
Family context

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

KnowledgeClaude 4.5 Sonnet (Reasoning) · 88
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningClaude 4.5 Sonnet (Reasoning) · 83
  1. 1
    GPT-6 Astra
    96
  2. 2
    Claude Mythos Preview
    95
  3. 3
    Gemini 3.7 Flash
    95
MathClaude 4.5 Sonnet (Reasoning) · 88
  1. 1
    OpenAI o3
    99
  2. 2
    GPT-5 (medium)
    99
  3. 3
    Grok 4
    99
CodingClaude 4.5 Sonnet (Reasoning) · 71
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticClaude 4.5 Sonnet (Reasoning) · 56
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextClaude 4.5 Sonnet (Reasoning) · 72
  1. 1
    Kimi K3
    89
  2. 2
    Claude Fable 5.1
    85
  3. 3
    GPT-5.5 (high)
    84
Newest receipts

Benchmark rows added to the public ledger for Claude 4.5 Sonnet (Reasoning) in the last 120 days. Older rows live in the full table below.

13 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context72.33%Artificial Analysis4
AA output speedPerformance0 t/sArtificial Analysis4
AA time to first tokenPerformance0.00sArtificial Analysis4
AIMEMath88%Artificial Analysis4
Artificial Analysis Intelligence IndexReasoning21.2%Artificial Analysis4
GPQA DiamondReasoning83.4%Artificial Analysis4
Humanity's Last ExamReasoning17.8%Artificial Analysis4
IFBenchReasoning57.28%Artificial Analysis4
LiveCodeBenchCoding71.4%Artificial Analysis4
MMLU-ProKnowledge87.5%Artificial Analysis4
SciCodeCoding45.7%Artificial Analysis4
τ³-Bench BankingAgentic24.54%Artificial Analysis4
Terminal-BenchAgentic55.81%Artificial Analysis4
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.

Knowledge88
Reasoning83
Math88
Coding71
Agentic56
Long context72
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.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaxAIMiniMaxZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Claude 4.5 Sonnet (Reasoning). 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 4.5 Sonnet (Reasoning).

Knowledge

Reasoning

Math

Coding

Agentic

Long context

Performance

Sources