Artificial intelligence · 2025

Claude 3.7 Sonnet

The first hybrid model: one model answers instantly or thinks at length.

Released February 24, 2025, Claude 3.7 Sonnet added an extended thinking mode with a user-set thinking budget, and scored 62.3% on SWE-bench Verified. It became the default model in many coding tools.

See it in the living tree Wikipedia

Arcs above join it to what it stood on, arcs below to what it led to. The bar places this stretch on the tree's whole timeline.

What it built on 2

Claude 3.5 Sonnet 2024
Reasoning models (o1) 2024

What it led to 2

Claude 4 2025
Claude Code 2025

Watch

Claude 3.7 Sonnet with extended thinkingAnthropic · YouTube

Related measurements

The dashed line marks when it first worked.

Best AI score on GPQA Diamond Latest: 95.8 % correct (2026) Source: Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv); Anthropic: Claude 3 public launch (March 4, 2024)

Training compute of notable AI models Source: Epoch AI, Notable AI Models (CC BY 4.0)