# Claude 3 Sonnet

> The original balanced Sonnet that launched the Claude 3 Opus/Sonnet/Haiku family.

- Maker: Anthropic
- Line: Claude Sonnet
- Canonical page: https://ai-tldr.dev/models/claude-3-sonnet/
- Updated: 2026-08-20

## In short

Claude 3 Sonnet is the original Sonnet model, announced on March 4, 2024 as the balanced middle tier of the Claude 3 family, between Claude 3 Opus and Claude 3 Haiku. 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 and Amazon Bedrock.

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | 200K (max output 4K) | [1] |
| Input price | $3.00 per million tokens | [2] |
| Output price | $15.00 per million tokens | [2] |
| Licence | Proprietary · API only | [1] |
| Released | 4 March 2024 | [1] |
| Knowledge cutoff | August 2023 | [1] |

## Sources

1. [The Claude 3 Model Family (model card PDF)](https://www-cdn.anthropic.com/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627/Model_Card_Claude_3.pdf)
2. [Introducing the next generation of Claude (Claude 3)](https://www.anthropic.com/news/claude-3-family)

## Overview

Claude 3 Sonnet is the original Sonnet model, announced on March 4, 2024 as the balanced middle tier of the Claude 3 family, between Claude 3 Opus and Claude 3 Haiku. It introduced the Opus/Sonnet/Haiku naming that every later Claude generation has used.

It was the first Claude model with vision, able to process photos, charts, graphs and technical diagrams alongside text. On standard benchmarks it scored 79.0% on MMLU, 40.4% on GPQA, 73.0% on HumanEval, and 92.3% on GSM8K, positioning it as a strong, affordable model for scaled enterprise deployments.

Claude 3 Sonnet has a 200K token context window, up to 4,096 output tokens, an August 2023 knowledge cutoff, and supports text and image input, at $3 / $15 per million tokens. Its model ID is claude-3-sonnet-20240229; the snapshot was retired on July 21, 2025.

## Benchmarks

- MMLU: 79% ([source](https://www.anthropic.com/news/claude-3-family))
- GPQA: 40.4% ([source](https://www.anthropic.com/news/claude-3-family))
- HumanEval: 73% ([source](https://www.anthropic.com/news/claude-3-family))
- GSM8K: 92.3% ([source](https://www.anthropic.com/news/claude-3-family))

## Pricing

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

Source: https://www.anthropic.com/news/claude-3-family

## Strengths

- Introduced the Opus/Sonnet/Haiku tier naming for the Claude family
- First Claude model with vision (charts, diagrams, photos)
- 200K token context window at an affordable mid-tier price
- Balanced intelligence and speed for enterprise scale
- Solid math and coding scores (92.3% GSM8K, 73.0% HumanEval)

## Best for

- General-purpose text generation and summarization
- Vision tasks over charts, graphs and technical diagrams
- Knowledge-base Q&A and enterprise content workflows
- Cost-sensitive deployments needing a balanced model
- Math and coding assistance at moderate difficulty

## How to access

- [Anthropic](https://www.anthropic.com/news/claude-3-family) — model id `claude-3-sonnet-20240229`
- [Amazon Bedrock](https://platform.claude.com/docs/en/build-with-claude/claude-in-amazon-bedrock) — model id `anthropic.claude-3-sonnet-20240229-v1:0`

## 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](https://ai-tldr.dev/models/claude-sonnet-4/) (22 May 2025)
- [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 (4 March 2024) — this page

## FAQ

### When was Claude 3 Sonnet released?

It was announced on March 4, 2024 as the balanced middle tier of the Claude 3 family (Opus, Sonnet, Haiku).

### What were its benchmark scores?

79.0% on MMLU, 40.4% on GPQA, 73.0% on HumanEval, and 92.3% on GSM8K.

### What are its key specs?

200K token context window, up to 4,096 output tokens, an August 2023 knowledge cutoff, text and vision input, at $3/$15 per million tokens.

### Is Claude 3 Sonnet still available?

No. The claude-3-sonnet-20240229 snapshot was retired on July 21, 2025. Anthropic recommends Claude Sonnet 4.6 as the replacement.

## Links

- [Introducing the next generation of Claude (Claude 3)](https://www.anthropic.com/news/claude-3-family)
- [The Claude 3 Model Family (model card PDF)](https://www-cdn.anthropic.com/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627/Model_Card_Claude_3.pdf)
- [Claude 3 Sonnet model card (PromptHub)](https://www.prompthub.us/models/claude-3-sonnet)
- [Anthropic model deprecations (retirement dates)](https://platform.claude.com/docs/en/about-claude/model-deprecations)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/claude-3-sonnet/