AnthropicClaude 4Arena #53May 22, 2025

Claude Opus 4

Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows. It sets new benchmarks in software engineering, achieving leading results on SWE-bench (72.5%) and Terminal-bench (43.2%). Opus 4 supports extended, agentic workflows, handling thousands of task steps continuously for hours without degradation. Read more at the blog post here

Context Window
200K
tokens
Max Output
32K
tokens
Released
May 22, 2025
Arena Rank
#53
of 305 models
Output Speed
68
tokens/sec
Time to First Token
1.8s
TTFT

Capabilities

👁Vision
🧠Reasoning
🔧Tool Calling
Prompt Caching
🖥Computer Use
🎨Image Generation

Supported Parameters

Include Reasoning
Show reasoning tokens
Max Tokens
Output length limit
Reasoning
Extended thinking
Stop Sequences
Custom stop tokens
Temperature
Controls randomness
Tool Choice
Control tool usage
Tool Calling
Function calling support
Top K
Top-K sampling
Top P
Nucleus sampling

Pricing Comparison

RouterInput / 1MOutput / 1MCached Input / 1M
Requesty$15.00$75.00$1.50
OpenRouter$15.00$75.00$1.50
Vercel AI$15.00$75.00
Martian$15.00$75.00$1.50

Benchmarks

Artificial Analysis
Intelligence IndexArtificial Analysis
72/100
Coding IndexArtificial Analysis
68/100
Math IndexArtificial Analysis
80/100
MMLU-PROArtificial Analysis
0.795/1
GPQA DiamondArtificial Analysis
0.715/1
MATH-500Artificial Analysis
0.905/1
AIME 2024Artificial Analysis
0.55/1
Humanity's Last ExamArtificial Analysis
0.112/1
LiveCodeBenchArtificial Analysis
0.685/1
SciCodeArtificial Analysis
0.395/1

Model IDs

Requestybedrock/claude-opus-4@us-east-1
OpenRouteranthropic/claude-opus-4

Tags

visionreasoningtool-callingcachingcomputer-use

Available Regions

EUUS
Compare with another model

Related Models