Claude Sonnet 4.5

Language modelPrevious generationCard summarized

The strongest model for building complex agents

Deprecated · retires 30 Nov 2026● Re-checked 12 Oct 2026 · 111 facts · 0 corrections · 4 unconfirmedSource: System Card: Claude Sonnet 4.5 (149 pp)

Use it via API

$1,764 /month

$35.28 per 1K requests

RAG over docs: 12K in · 600 out · 50K requests/mo · 30% cached · change

Code example Compare

Where to call it

Claude APIclaude-sonnet-4-5-20250929
Amazon Bedrockanthropic.claude-sonnet-4-5-20250929-v1:0
Google Vertex AIclaude-sonnet-4-5@20250929
Microsoft Foundry / Azureclaude-sonnet-4-5

Supports

Image input
PDF input
Structured outputs
Reasoning control
Tool calling
Web search
Prompt caching
Batch API

Full feature support

Good for

Software coding
Agentic tasks where the model runs in a loop and uses tools
Computer use

Replaced by

Claude Sonnet 5.5$2.00 / $10.00

This model: $3.00 / $15.00 per 1M tokens

Compare side by side

Overview

Anthropic · 14Sonnet

Sonnet 4.5

The strongest model for building complex agents

Cost: $3.00 in
Context: 200K

Input / output: $3.00 / $15.00
Max output: 64K
Knows up to: January 2025
Released: 29 Sep 2025

A hybrid reasoning large language model from Anthropic with strengths in coding, agentic tasks and computer use. Anthropic finds it has a substantially improved safety profile compared to previous Claude models, and deployed it under the AI Safety Level 3 Standard. p.3

claude-sonnet-4-5-20250929

Input / output per 1M: $3.00 / $15.00
Context window: 200K
Max output: 64K
Knows up to: January 2025
Reasoning: Extended
Weights: Closed

About this card. A model card is the maker's own account of a model: what it's for, how it was trained and tested, and where it falls short. This page condenses Anthropic's System Card: Claude Sonnet 4.5. Every claim carries its source page.