Claude Sonnet 4

Retired · Language, Multimodal, Reasoning · Milestone

Claude Sonnet 4 is a language model in Anthropic's Sonnet line that accepts text and images, released together with Claude Opus 4. It was offered to free and paid Claude users, on the Anthropic API, Amazon Bedrock and Google Cloud Vertex AI, with near-instant answers and an extended thinking mode, and Anthropic presents it as an upgrade to Claude 3.7 Sonnet. [1] [2]Claude Opus 4 and Claude Sonnet 4 Summary Table, Modalities Primary source

Timeline of Claude Sonnet 4 →

Claims and evidence

  • Released Primary source[1] [3]May 22, 2025
  • Deprecated Primary source[4]2026-04-14: Claude Sonnet 4 and Claude Opus 4 models [3]April 14, 2026
  • Retired Primary source[4]2026-04-14: Claude Sonnet 4 and Claude Opus 4 models [3]June 15, 2026
  • Status Retired Primary source[4]Model status
  • Replaced by Claude Sonnet 4.6Recommended replacement as listed on the deprecations page when checked on 2026-10-01. Primary source[4]2026-04-14: Claude Sonnet 4 and Claude Opus 4 models
  • Change · Tool use Can use tools such as web search during extended thinking; this was in beta at launch.Compared with Claude 3.7 Sonnet Primary source[1]
  • Change · Reasoning Long thought processes are condensed by a smaller model, where Claude 3.7 Sonnet always showed its raw thinking; Anthropic says this is needed in about 5% of cases.Compared with Claude 3.7 Sonnet Primary source[5] [1]
  • Change · Training data Training data includes public web data as of March 2025, against November 2024 for Claude 3.7 Sonnet.Compared with Claude 3.7 Sonnet Primary source[5] [6]
  • Change · Context length A 1M-token context window was offered in beta from 2025-08-12, after launch; Anthropic ended that beta for this model on 2026-04-30.Compared with Claude 3.7 Sonnet Primary source[3]August 12, 2025; March 30, 2026; April 30, 2026
  • Input text, image Primary source[2]Claude Opus 4 and Claude Sonnet 4 Summary Table, Modalities
  • Output textThe model report also mentions audio output through text-to-speech in the Claude apps. Primary source[2]Claude Opus 4 and Claude Sonnet 4 Summary Table, Modalities
  • Feature Reasoning mode Primary source[1] [5]section 1.1.2
  • Context window 200k tokensA 1M-token window was in beta from 2025-08-12 and was retired for this model on 2026-04-30. Primary source[3]March 30, 2026
  • Parameters Not disclosedNot disclosed by Anthropic. No claim madeNo source
  • Access consumer app, API, cloud partner Primary source[1]

Lineage

Predecessors

Successors

Based on

Not derived from another model.

Variants and derived

None recorded.

Siblings

None recorded.

All ancestors

All descendants

Variants

No variants recorded in this record.

Related AI Radar coverage

AI Radar coverage starts in June 2026; no coverage linked yet.

All model releases from Anthropic on AI Radar →