Claude Opus 4
Multimodal
Zero-eval
#1TAU-bench Retail
#1MMMU (validation)
#1Terminal-bench
+4 more
by Anthropic
About
Claude Opus 4 is a multimodal language model developed by Anthropic. It achieves strong performance with an average score of 64.6% across 9 benchmarks. It excels particularly in MMMLU (88.8%), TAU-bench Retail (81.4%), GPQA (79.6%). It supports a 328K token context window for handling large documents. The model is available through 3 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
Providers3
Timeline
AnnouncedMay 22, 2025
ReleasedMay 22, 2025
Specifications
Capabilities
Multimodal
License & Family
License
Proprietary
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
9 benchmarks
Average Score
64.6%
Best Score
88.8%
High Performers (80%+)
2Performance Metrics
Max Context Window
328.0KAvg Throughput
87.3 tok/sAvg Latency
0msTop Categories
vision
76.5%
general
71.1%
agents
70.5%
reasoning
8.6%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
MMMLU
Rank #2 of 13
#1Claude Opus 4.1
98.4%
#2Claude Opus 4
88.8%
#3o1
87.7%
#4GPT-4.1
87.3%
#5Qwen3 235B A22B
86.7%
TAU-bench Retail
Rank #1 of 15
#1Claude Opus 4
81.4%
#2Claude 3.7 Sonnet
81.2%
#3Claude Sonnet 4
80.5%
#4o4-mini
71.8%
GPQA
Rank #14 of 115
#11DeepSeek-R1-0528
81.0%
#12o4-mini
81.4%
#13GPT-5 mini
82.3%
#14Claude Opus 4
79.6%
#15o1-pro
79.0%
#16o1
78.0%
#17Qwen3-235B-A22B-Instruct-2507
77.5%
MMMU (validation)
Rank #1 of 2
#1Claude Opus 4
76.5%
#2Claude Opus 4.1
64.8%
AIME 2025
Rank #16 of 36
#13Phi 4 Reasoning Plus
78.0%
#14Claude Opus 4.1
80.2%
#15Qwen3 235B A22B
81.5%
#16Claude Opus 4
75.5%
#17Qwen3 32B
72.9%
#18Llama 3.1 Nemotron Ultra 253B v1
72.5%
#19Gemini 2.5 Flash
72.0%
All Benchmark Results for Claude Opus 4
Complete list of benchmark scores with detailed information
MMMLU MMMLU benchmark | general | text | 0.89 | 88.8% | Self-reported |
TAU-bench Retail TAU-bench Retail benchmark | agents | text | 0.81 | 81.4% | Self-reported |
GPQA GPQA benchmark | general | text | 0.80 | 79.6% | Self-reported |
MMMU (validation) MMMU (validation) benchmark | vision | multimodal | 0.77 | 76.5% | Self-reported |
AIME 2025 AIME 2025 benchmark | general | text | 0.76 | 75.5% | Self-reported |
SWE-Bench Verified SWE-Bench Verified benchmark | general | text | 0.72 | 72.5% | Self-reported |
TAU-bench Airline TAU-bench Airline benchmark | agents | text | 0.60 | 59.6% | Self-reported |
Terminal-bench Terminal-bench benchmark | general | text | 0.39 | 39.2% | Self-reported |
ARC-AGI v2 ARC-AGI v2 benchmark | reasoning | text | 0.09 | 8.6% | Unverified |
Resources