Claude Sonnet 4.5
Multimodal
Zero-eval
#1Tau2 Telecom
#1Tau2 Retail
#1MMMUval
+5 more
by Anthropic
About
Claude Sonnet 4.5 is a multimodal language model developed by Anthropic. It achieves strong performance with an average score of 78.0% across 10 benchmarks. It excels particularly in Tau2 Telecom (98.0%), MMMLU (89.1%), AIME 2025 (87.0%). It supports a 264K token context window for handling large documents. The model is available through 2 API providers. As a multimodal model, it can process and understand text, images, and other input formats seamlessly. Released in 2025, it represents Anthropic's latest advancement in AI technology.
Pricing Range
Input (per 1M)$3.00 -$3.00
Output (per 1M)$15.00 -$15.00
Providers2
Timeline
AnnouncedSep 29, 2025
ReleasedSep 29, 2025
Knowledge CutoffJan 31, 2025
Specifications
Capabilities
Multimodal
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown
Overall Performance
10 benchmarks
Average Score
78.0%
Best Score
98.0%
High Performers (80%+)
5Performance Metrics
Max Context Window
264.0KAvg Throughput
42.0 tok/sAvg Latency
0msTop Categories
Math
87.0%
Agents
82.8%
General
78.0%
Vision
77.8%
Code
50.0%
All Benchmark Results for Claude Sonnet 4.5
Complete list of benchmark scores with detailed information
| Tau2 Telecom | agents | text | 0.98 | 98.0% | Self-reported |
| MMMLU | general | text | 0.89 | 89.1% | Self-reported |
| AIME 2025 | math | text | 0.87 | 87.0% | Self-reported |
| Tau2 Retail | agents | text | 0.86 | 86.2% | Self-reported |
| GPQA | general | text | 0.83 | 83.4% | Self-reported |
| MMMUval | vision | multimodal | 0.78 | 77.8% | Self-reported |
| SWE-bench Verified (Agentic Coding) | agents | text | 0.77 | 77.2% | Self-reported |
| Tau2 Airline | agents | text | 0.70 | 70.0% | Self-reported |
| OSWorld | general | text | 0.61 | 61.4% | Self-reported |
| Terminal-Bench | code | text | 0.50 | 50.0% | Self-reported |
Resources