Claude Opus 4.5
Claude Opus 4.5 is Anthropic's flagship AI model released November 24, 2025. The first model to exceed 80% on SWE-bench Verified (80.9%). Features the effort parameter, extended thinking, computer use, and industry-leading prompt injection resistance.
Technical specification
Modalities
Capabilities
12Reasoning★
Reasoning
Multi-step reasoning★
Reasoning
Long context★
Reasoning
Coding★
Coding
Function Calling
Planning
Structured output★
Structured gen.
Image understanding★
Vision
Chart understanding
Vision
OCR★
Vision
Multilingual★
Language
Planning★
Planning
Streaming output
Reasoning
Applications
$5 per million input tokens, $25 per million output tokens. Approximately 67% reduction compared to Opus 4.1 ($15/$75). Prompt caching: write at 5 min — 1.25× base price, read — 0.1× (up to 90% savings). Batch API: 50% discount; processing within 24 hours.
INPUT
$5.0000 / per 1M tokens
OUTPUT
$25.0000 / per 1M tokens
CACHE
$0.5000 / per 1M tokens
TOTAL
for 10K tokens
Standard text tokens. Cache read = 0.50 USD/MTok (10% of base price).
50% discount with asynchronous Batch API (processing within 24 h).
Security and enterprise
Claude Opus 4.5 uses the publicly documented platform-level security controls provided by Anthropic. Security information applies primarily to Claude as a product, the Anthropic API, and enterprise features, rather than a separate security profile specific to the Opus 4.5 version.
Technical information
In practice, the security of Opus 4.5 should be treated as inherited security from the Anthropic platform and enterprise controls.