Claude Sonnet 4.5 is Anthropic's most advanced Sonnet model prior to the 4.6 generation, optimized for real-world agents and coding workflows. It delivers state-of-the-art coding performance with improvements in system design, code security, and specification adherence. The model is designed for extended autonomous operation with fact-based progress tracking, and introduces stronger agentic capabilities including speculative parallel execution, improved tool orchestration, and efficient context and memory management for multi-context, long-running workflows.
Anthropic FreeAnthropic ProAnthropic Max (5x)Anthropic Max (20x)API|VisionReasoningWeb SearchFile|Proprietary Model
Knowledge Cutoff
2025-01-31
Input → Output Format
Context Memory
1MIN64KOUT
AI Performance Evaluation
Arena Overall Score
1453
±3As of 2026-04-23
Overall Rank
No.30
63,725 Votes
Arena by Ability
Hard Prompts
1485±4No.22
Expert Knowledge
1498±10No.15
Instruction Following
1463±5No.14
Conversation Memory
1469±6No.27
Creative
1442±7No.24
Coding
1519±6No.14
Math
1454±10No.28
Arena by Occupation
Creative Writing
1446±6No.23
Social Sciences
1467±6No.34
Media
1438±6No.22
Business
1454±6No.30
Healthcare
1465±10No.47
Legal
1458±10No.41
Software
1498±5No.22
Mathematics
1467±11No.22
Source:Arena Intelligence
Overall
AA Intelligence Index
43%↑5%
LiveBench
51%↓9%
ForecastBench
60%↑1%
Reasoning & Math
AA Math Index
88%↑15%
GPQA Diamond
83%↑2%
HLE
17%↑0%
MMLU-Pro
88%↑6%
AIME 2025
88%↑14%
LB Reasoning
42%↓17%
LB Math
63%↓11%
LB Data
47%↓3%
Coding
AA Coding Index
39%↑5%
LiveCodeBench
71%↑6%
LB Coding
76%↑2%
LB Agentic
48%↑5%
TAU2
78%↑5%
TerminalBench
36%↑5%
SciCode
45%↑4%
Language & Instructions
IFBench
57%↑1%
AA-LCR
66%↑4%
Hallucination (HHEM)
12%↑2%
Factual (HHEM)
88%↓2%
LB Language
76%↑4%
LB IF
24%↓23%
Output Speed
Standard Mode
44tok/s↓38
First Output 0.80s
Reasoning Mode
60tok/s↓28
First Output 8.01s
