Codesota · Models · Claude Sonnet 4Anthropic20 results · 14 benchmarks
Model card

Claude Sonnet 4.

AnthropicapiMultimodal LLMProprietary1 current SOTA

#1 on ThaiOCRBench

§ 01 · Benchmarks

Every benchmark Claude Sonnet 4 has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01CrossCodeEvalComputer Code · Code Completionexact-match44.5%#1/6source ↗
02ThaiOCRBenchComputer Vision · Optical Character Recognitionted-score0.8%#1/5source ↗
03TransCoder (GeeksForGeeks)Computer Code · Code Translationcomputational-accuracy89.4%#1/7source ↗
04MBPPComputer Code · Code Generationpass@189.6%#6/19source ↗
05GSM8KReasoning · Mathematical Reasoningaccuracy97.8%#12/32source ↗
06MMLUReasoning · Commonsense Reasoningaccuracy90.1%#14/41source ↗
07HumanEvalComputer Code · Code Generationpass@190.6%#16/42source ↗
08SWE-Bench VerifiedComputer Code · Code Generationresolve-rate72.7%#16/39source ↗
09OCRBench v2Computer Vision · General OCR Capabilitiesoverall-en-private42.4%#17/272025-05-22source ↗
10LiveCodeBenchComputer Code · Code Generationpass@152.8%#18/302024-03-12source ↗
11GPQAReasoning · Multi-step Reasoningaccuracy70.0%#18/33source ↗
12MATHReasoning · Mathematical Reasoningaccuracy88.9%#21/34source ↗
13SWE-bench VerifiedAgentic AI · SWE-benchresolve-rate72.7%#32/81source ↗
14PLCCNatural Language Processing · Polish Cultural Competencyhistory81.0%#60/165source ↗
15PLCCNatural Language Processing · Polish Cultural Competencyculture-and-tradition72.0%#64/165source ↗
16PLCCNatural Language Processing · Polish Cultural Competencyvocabulary61.0%#66/165source ↗
17PLCCNatural Language Processing · Polish Cultural Competencyaverage68.2%#69/165source ↗
18PLCCNatural Language Processing · Polish Cultural Competencygeography77.0%#71/165source ↗
19PLCCNatural Language Processing · Polish Cultural Competencygrammar63.0%#74/165source ↗
20PLCCNatural Language Processing · Polish Cultural Competencyart-and-entertainment55.0%#77/165source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area

Where Claude Sonnet 4 actually performs.

Computer Vision
2
benchmarks
avg rank #9.0 · 1 SOTA
Computer Code
6
benchmarks
avg rank #9.7
Reasoning
4
benchmarks
avg rank #16.3
Agentic AI
1
benchmark
avg rank #32.0
Natural Language Processing
1
benchmark
avg rank #68.7
§ 03 · Papers

1 paper with results for Claude Sonnet 4.

  1. 2024-03-12· Computer Code· 1 result

    LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code

§ 04 · Related models

Other Anthropic models scored on Codesota.

Claude Opus 4
Undisclosed params · 13 results · 2 SOTA
Claude Opus 4.5
3 results · 2 SOTA
Claude Sonnet 5
Undisclosed params · 2 results · 2 SOTA
Claude Mythos Preview
1 result · 1 SOTA
Claude 3.5 Sonnet
Undisclosed params · 27 results
Claude Opus 4.5
Undisclosed params · 13 results
Claude 3.7 Sonnet
10 results
Claude 3 Opus
5 results
§ 05 · Sources & freshness

Where these numbers come from.

sdadas/PLCC
7
results
official-model-card
4
results
anthropic-model-card
4
results
alphaxiv-leaderboard
1
result
anthropic-blog
1
result
ocrbench-v2-leaderboard
1
result
official-leaderboard
1
result
editorial
1
result
16 of 20 rows marked verified. · first result 2024-03-12, latest 2025-05-22.