Claude 3 Sonnet
Claude 3 Sonnet is a closed AI model from Anthropic with an index score of 98 (GPT-4 = 100), rank 200 of 217.
9881–115Index score#200 of 217 models
–Price per million tokensno public API price found
–Context windowtokens it reads at once
4 Mar 2024ReleasedReplaced by Claude 3.5 Sonnet (Jun 2024)
Where it sitson the frontier
GPT-4 reached this score 12 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workno evidence yet
Novel problemsno evidence yet
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 9 leaderboards
The Model Indexthird-party tests, our method#200of 217 · 98
LMArena · Codingcoding questions in chat#244of 357 · 1318LMArena · Instruction followingfollowing detailed instructions#248of 362 · 1268LMArena · Creative writingcreative writing#248of 360 · 1246LMArena · Textoverall chat quality, judged by people#252of 362 · 1281LMArena · Mathsmaths questions in chat#242of 346 · 1254Epoch Capabilities Indexoverall capability from many benchmarks#163of 231 · 119.7LMArena · Long promptslong, complex requests#248of 340 · 1288LMArena · Visionunderstanding images#132of 145 · 1017Best attop three among current models
Replaced by Claude 3.5 Sonnet (Jun 2024); it no longer leads any category among current models.
Where to use itprices checked 18 Nov 2026
Host prices from OpenRouter; going direct can differ.
Facts and sources
Released4 Mar 2024
Announcementanthropic.com ↗
Release entryClaude 3 (Opus/Sonnet; Haiku 03-13)
Scored asClaude 3 Sonnet
Output speed–
Hugging Faceclosed weights