Back to Models
AnthropicProprietary
Anthropic Claude Opus 4.6
Claude Opus 4.6, developed by Anthropic, is an inference model with advanced reasoning abilities. It is a closed-source foundation model equipped with an extremely long context window of 1000K.
Parameters
Undisclosed
Context Window
1000K
License
Proprietary
Release Date
2026-02-05
API Pricing
Input Price (per 1M tokens)
$2.5
Output Price (per 1M tokens)
$
Billing Mode: standard
Strengths
- ・Provides advanced reasoning capabilities
- ・Massive 1000K context grasp
- ・Handles complex tasks effectively
Weaknesses
- ・Non-public and opaque model
- ・Commercial license restrictions
- ・Not open-source
Use Cases
- ・Large-scale document analysis
- ・Tasks requiring complex logical thinking
- ・Long-context processing
Deep Analysis
Arena Elo
1498
#1 overall
SWE-Bench Verified
80.8%
Production frontier
GPQA Diamond
91.3%
Pricing
$5.00/$25.00 per 1M tokens
Context Window
1M tokens
Release Date
February 5, 2026
Strengths
- ・#1 on Arena Elo (1498)
- ・SWE-Bench 80.8% — highest production score
- ・GPQA 91.3% leads reasoning benchmarks
- ・1M context window
- ・Strong instruction following
Weaknesses
- ・Expensive at $5/$25 per 1M tokens
- ・Slower than Haiku/Sonnet for latency-sensitive tasks
- ・Text+image input only, no audio/video
Competitor Comparison
| Model | Arena | SWE | GPQA | Price |
|---|---|---|---|---|
| GPT-5.4 | ~1470 | 77.5% | 92.0% | $2.50/$10.00 |
| Gemini 3 Pro | ~1449 | 76.2% | $1.25/$5.00 | |
| Claude Sonnet 5 | ~1430 | 92.4% | $3/$15 |
Claude Opus 4.6 is Anthropic's flagship model and the highest-ranked on Chatbot Arena (Elo 1498). Leads on SWE-Bench Verified (80.8%) and GPQA Diamond (91.3%) with a 1M token context window.
Sources
Analysis generated: 2026-05-24