# Claude Sonnet 4

> The Claude 4 generation's balanced model, with state-of-the-art coding at launch.

- Maker: Anthropic
- Line: Claude Sonnet
- Canonical page: https://ai-tldr.dev/models/claude-sonnet-4/
- Updated: 2026-06-20

## In short

Claude Sonnet 4 is the Claude 4 generation's balanced model, with state-of-the-art coding at launch. Anthropic released it on 22 May 2025 as an API-only model. It has a 200K-token context window and costs $3.00 per million input tokens and $15.00 per million output tokens. You can use it through Anthropic, Amazon Bedrock and Google Vertex AI.

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | 200K (max output 64K) | [1] |
| Input price | $3.00 per million tokens | [2] |
| Output price | $15.00 per million tokens | [2] |
| Licence | Proprietary · API only | [1] |
| Released | 22 May 2025 | [1] |
| Knowledge cutoff | March 2025 | [1] |

## Sources

1. [Models overview (Claude docs)](https://platform.claude.com/docs/en/about-claude/models/overview)
2. [Introducing Claude 4](https://www.anthropic.com/news/claude-4)

## Overview

Claude Sonnet 4 was released on May 22, 2025, alongside Claude Opus 4 as part of the Claude 4 model generation. It is a hybrid reasoning model that significantly improved on Claude 3.7 Sonnet's coding, reasoning, and instruction-following, reaching a state-of-the-art 72.7% on SWE-bench Verified at launch.

Sonnet 4 introduced the Claude 4 generation's improvements in steerability and agentic behaviour, with extended thinking available for harder math and science problems. It scored 70.0% on GPQA Diamond and 33.1% on AIME without extended thinking, and was offered to free users as well as on the paid plans.

Sonnet 4 has a 200K token context window, 64K max output tokens, and supports text and vision input. It was offered on the Claude API as claude-sonnet-4-0 (snapshot claude-sonnet-4-20250514), Amazon Bedrock, and Vertex AI, at $3 / $15 per million tokens. The snapshot was retired on June 15, 2026 — Anthropic recommends migrating to Claude Sonnet 4.6.

## Benchmarks

- SWE-bench Verified: 72.7% ([source](https://www.anthropic.com/news/claude-4))
- GPQA Diamond: 70% ([source](https://www.anthropic.com/news/claude-4))
- AIME (no extended thinking): 33.1% ([source](https://www.anthropic.com/news/claude-4))

## Pricing

- Input: $3.00 per million tokens
- Output: $15.00 per million tokens

Same pricing as Claude 3.7 Sonnet.

Source: https://www.anthropic.com/news/claude-4

## Strengths

- State-of-the-art 72.7% SWE-bench Verified at launch
- Large coding and steerability gains over Claude 3.7 Sonnet
- Extended thinking for math, science and multi-step reasoning
- Available to free users at launch
- Same $3/$15 pricing as prior Sonnet generations

## Best for

- Agentic coding and software-engineering tasks
- General reasoning and instruction-following assistants
- Multimodal document and image understanding
- Tool-using agents and workflow automation
- Cost-efficient alternative to Opus 4 for everyday work

## How to access

- [Anthropic](https://platform.claude.com/docs/en/about-claude/models/overview) — model id `claude-sonnet-4-0`
- [Amazon Bedrock](https://platform.claude.com/docs/en/build-with-claude/claude-in-amazon-bedrock) — model id `anthropic.claude-sonnet-4-20250514-v1:0`
- [Google Vertex AI](https://platform.claude.com/docs/en/build-with-claude/claude-on-vertex-ai) — model id `claude-sonnet-4@20250514`

## Claude Sonnet versions

- [Claude Sonnet 4.6](https://ai-tldr.dev/models/claude-sonnet-4-6/) (17 February 2026)
- [Claude Sonnet 4.5](https://ai-tldr.dev/models/claude-sonnet-4-5/) (29 September 2025)
- Claude Sonnet 4 (22 May 2025) — this page
- [Claude 3.7 Sonnet](https://ai-tldr.dev/models/claude-3-7-sonnet/) (24 February 2025)
- [Claude 3.5 Sonnet (new)](https://ai-tldr.dev/models/claude-3-5-sonnet-v2/) (22 October 2024)
- [Claude 3.5 Sonnet](https://ai-tldr.dev/models/claude-3-5-sonnet/) (20 June 2024)
- [Claude 3 Sonnet](https://ai-tldr.dev/models/claude-3-sonnet/) (4 March 2024)

## FAQ

### When was Claude Sonnet 4 released?

May 22, 2025, alongside Claude Opus 4 as part of the Claude 4 generation.

### How did Sonnet 4 score on coding benchmarks?

It reached a state-of-the-art 72.7% on SWE-bench Verified at launch, a large improvement over Claude 3.7 Sonnet.

### What is the context window and pricing?

200K token context, 64K max output, priced at $3 per million input and $15 per million output tokens.

### Is Claude Sonnet 4 still recommended?

No — the claude-sonnet-4-20250514 snapshot was retired on June 15, 2026 and now returns an error. Anthropic recommends migrating to Claude Sonnet 4.6, which offers far stronger coding and computer-use performance at the same price.

## Links

- [Introducing Claude 4](https://www.anthropic.com/news/claude-4)
- [Models overview (Claude docs)](https://platform.claude.com/docs/en/about-claude/models/overview)
- [Claude Sonnet 4 benchmarks (LLM-Stats)](https://llm-stats.com/models/claude-sonnet-4-20250514)
- [DataCamp: Claude 4 review](https://www.datacamp.com/blog/claude-4)
- [Claude API release notes — Sonnet 4 retirement (June 15, 2026)](https://platform.claude.com/docs/en/release-notes/api)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/claude-sonnet-4/