DataLearner logoDataLearnerAI
Latest AI Insights
Model Leaderboards
Benchmarks
Model Directory
Model Comparison
Resource Center
Tools
LanguageEnglish
DataLearner logoDataLearner AI

A knowledge platform focused on LLM benchmarking, datasets, and practical instruction with continuously updated capability maps.

Products

  • Leaderboards
  • Model comparison
  • Datasets

Resources

  • Tutorials
  • Editorial
  • Tool directory

Company

  • About
  • Privacy policy
  • Data methodology
  • Contact

© 2026 DataLearner AI. DataLearner curates industry data and case studies so researchers, enterprises, and developers can rely on trustworthy intelligence.

Privacy policyTerms of service
Page navigation
Page navigation
Model catalogClaude Sonnet 4.5
CL

Claude Sonnet 4.5

AI model

Claude Sonnet 4.5

Release date: 2025-09-30Updated: 2025-10-19 12:28:123,903
Live demoGitHubHugging FaceCompare
Parameters
Not disclosed
Context length
1000K
Chinese support
Supported
Reasoning ability

Claude Sonnet 4.5 is an AI model published by Anthropic, released on 2025-09-30, for AI model, and 1000K tokens context length, with no open-source license.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Claude Sonnet 4.5

Model basics

Reasoning traces
Supported
Thinking modes
Thinking modes not supported
Context length
1000K tokens
Max output length
65536 tokens
Model type
AI model
Release date
2025-09-30
Model file size
No data
MoE architecture
No
Total params / Active params
No data / N/A
Knowledge cutoff
No data
Claude Sonnet 4.5

Open source & experience

Code license
Not open source
Weights license
Not open source
GitHub repo
GitHub link unavailable
Hugging Face
Hugging Face link unavailable
Live demo
https://claude.ai/
Claude Sonnet 4.5

Official resources

Paper
Introducing Claude Sonnet 4.5
DataLearnerAI blog
DataLearnerAI blog
Claude Sonnet 4.5

API details

API speed
3/5
💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.
Standard pricingStandard
ModalityInputOutput
Text$3$15
Cached pricingCache
ModalityInput cacheOutput cache
Text$3.75$0.3
Extended context pricingExtended
ModalityInputOutput
Text$6$22.5
Claude Sonnet 4.5

Benchmark Results

Claude Sonnet 4.5 currently shows benchmark results led by AIME2025 (1 / 106, score 100), SWE-bench Verified (3 / 103, score 82), MMLU Pro (5 / 124, score 88). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking
Tool usage

General Knowledge

12 evaluations
Benchmark / mode
Score
Rank/total
MMLU Pro
Thinking Mode
88
5 / 124
GPQA Diamond
Standard Mode
73.70
94 / 175
GPQA Diamond
Thinking Mode
83.40
55 / 175
LiveBench
Standard Mode
70.56
20 / 52
LiveBench
Thinking Mode
78.26
4 / 52
ARC-AGI
Standard Mode
25.50
52 / 65
ARC-AGI
Thinking Mode
63.70
32 / 65
HLE
Standard Mode
7.10
136 / 149
HLE
33.60
60 / 149
HLE
Thinking Mode
17.70
103 / 149
ARC-AGI-2
Standard Mode
3.80
48 / 58
ARC-AGI-2
Thinking Mode
13.60
34 / 58

Coding and Software Engineer

5 evaluations
Benchmark / mode
Score
Rank/total
SWE-bench Verified
82
3 / 103
SWE-bench Verified
77.20
20 / 103
LiveCodeBench
Standard Mode
59
69 / 118
LiveCodeBench
Thinking Mode
71
45 / 118
SWE-Bench Pro - Public
Thinking Mode
43.60
29 / 36

Math and Reasoning

8 evaluations
Benchmark / mode
Score
Rank/total
AIME2025
Standard Mode
37
96 / 106
AIME2025
100
1 / 106
AIME2025
Thinking Mode
87
45 / 106
IMO-ProofBench
Thinking Mode
27.10
8 / 16
FrontierMath
Standard Mode
5.20
38 / 60
IMO-ProofBench Advanced
Thinking Mode
4.80
6 / 8
FrontierMath - Tier 4
Standard Mode
2.10
56 / 80
FrontierMath - Tier 4
32K
4.20
40 / 80

AI Agent - Tool Usage

4 evaluations
Benchmark / mode
Score
Rank/total
OSWorld-Verified
61.40
10 / 14
Terminal-Bench
50
3 / 35
Terminal-Bench
27
25 / 35
Terminal Bench 2.0
42.80
38 / 43

Multimodal Understanding

1 evaluations
Benchmark / mode
Score
Rank/total
MMMU
Thinking Mode
77.80
14 / 28

Math and Reasoning

1 evaluations
Benchmark / mode
Score
Rank/total
Simple Bench
Standard Mode
54.30
9 / 27

Agent Level Benchmark

4 evaluations
Benchmark / mode
Score
Rank/total
τ²-Bench - Telecom
98
5 / 35
τ²-Bench
84.70
9 / 40
τ²-Bench
71
24 / 40
Terminal Bench Hard
33
8 / 13

Instruction Following

1 evaluations
Benchmark / mode
Score
Rank/total
IF Bench
57.30
19 / 27

AI Agent - Information Search

1 evaluations
Benchmark / mode
Score
Rank/total
BrowseComp
24.10
41 / 43

Productivity Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
GDPval-AA
Thinking Mode
39
15 / 20

Long Context

1 evaluations
Benchmark / mode
Score
Rank/total
AA-LCR
Thinking Mode
66
8 / 13

Claw-style Agent Evaluation

2 evaluations
Benchmark / mode
Score
Rank/total
Pinch Bench
Thinking ModeTools
88.20
4 / 37
Claw Bench
Thinking ModeTools
88.10
13 / 29
View benchmark analysisCompare with other models

Compare with other models

  • Earlier versionClaude Sonnet 4.5 vs Claude Sonnet 425 benchmarks
  • Peer modelClaude Sonnet 4.5 vs Gemini 2.5-Pro23 benchmarks
  • Peer modelClaude Sonnet 4.5 vs GPT-5.115 benchmarks
  • Earlier versionClaude Sonnet 4.5 vs Claude Sonnet 3.713 benchmarks
  • Earlier versionClaude Sonnet 4.5 vs Claude 3.5 Sonnet New6 benchmarks
  • Earlier versionClaude Sonnet 4.5 vs Claude 3.5 Sonnet4 benchmarks

Want a custom combination? Open the compare tool

Claude Sonnet 4.5

Publisher

Anthropic
Anthropic
View publisher details
Claude Sonnet 4.5

Model Overview

Claude Sonnet 4.5 is an AI model published by Anthropic, released on 2025-09-30, for AI model, and 1000K tokens context length, with no open-source license.

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.

DataLearner WeChat QR code