Claude Opus 4.1
Multimodal
Zero-eval
#1MMMLU
#1MMMU (validation)
#2TAU-bench Retail
+2 more
by Anthropic
+
+
+
+
About
Claude Opus 4.1 is a multimodal language model developed by Anthropic. It achieves strong performance with an average score of 72.7% across 8 benchmarks. It excels particularly in MMMLU (89.5%), TAU-bench Retail (82.4%), GPQA (80.9%). It supports a 232K token context window for handling large documents. The model is available through 4 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)$15.00 -$15.00
Output (per 1M)$75.00 -$75.00
Providers4
+
+
+
+
Timeline
AnnouncedAug 5, 2025
ReleasedAug 5, 2025
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown
Overall Performance
8 benchmarks
Average Score
72.7%
Best Score
89.5%
High Performers (80%+)
3Performance Metrics
Max Context Window
232.0KAvg Throughput
76.0 tok/sAvg Latency
0ms+
+
+
+
All Benchmark Results for Claude Opus 4.1
Complete list of benchmark scores with detailed information
| MMMLU | text | 0.90 | 89.5% | Self-reported | |
| TAU-bench Retail | text | 0.82 | 82.4% | Self-reported | |
| GPQA | text | 0.81 | 80.9% | Self-reported | |
| AIME 2025 | text | 0.78 | 78.0% | Self-reported | |
| MMMU (validation) | multimodal | 0.77 | 77.1% | Self-reported | |
| SWE-Bench Verified | text | 0.74 | 74.5% | Self-reported | |
| TAU-bench Airline | text | 0.56 | 56.0% | Self-reported | |
| Terminal-Bench | text | 0.43 | 43.3% | Self-reported |