← Back to all models
C

Claude Opus 4.6

by Anthropic
Flagship Paid API Multimodal 🏆 Ranked #6 of 85
89.3
Overall Score
out of 100
About

Anthropic's hybrid reasoning flagship with a 1M context window, pushing the frontier for coding and agentic tasks. Combines extended thinking with tool use for complex multi-step workflows.

Key Metrics
Context Window
1.0M
tokens
Avg Response
1100
milliseconds
Input Cost
$18.0
per million tokens
Output Cost
$90.0
per million tokens
Arena ELO
1360
Chatbot Arena rating
MT-Bench
9.4
out of 10
Benchmark Scores
MMLU
92.0%
HumanEval
93.0%
MATH
90.0%
GPQA
75.0%
MT-Bench
94.0/10
Capability Profile
Strengths & Limitations
Strengths
✓ 1M context window ✓ Hybrid reasoning ✓ Coding excellence ✓ Agentic capabilities ✓ Long-range tasks
Limitations
⚠ Highest cost tier ⚠ Slower with extended thinking ⚠ Early-access pricing
Ideal Use Cases
Agentic AI Full codebase analysis Long document processing Complex coding Multi-step research
Model Details
Provider Anthropic
Released 2025-06-01
Type Paid API
Multimodal Yes
Tier Flagship
Global rank #6 / 85
Pricing (USD)
Input tokens $18.0/M
Output tokens $90.0/M
Per 1,000 tokens ≈ $0.0180 input / $0.0900 output
All Benchmarks
MMLU 92.0%
HumanEval 93.0%
MATH 90.0%
GPQA 75.0%
MT-Bench 9.4/10
Arena ELO 1360
Compare this model View Rankings

You might also consider