BenchAtlas

Rankings / Anthropic

Claude Sonnet 5

proprietary

released 2026-06-30

BenchAtlas Index

as of 2026-07-14
75.9
max effort
rank #31
7 families · 4 categories · medium
74.0
max effort
rank #42
6 families · 4 categories · medium
69.5
xhigh effort
rank #67
8 families · 5 categories · high

Benchmark evidence

56 results
agentic coding
Release 2026-06-25
tasks_counted=3 · livebench_version=2026-06-25
xhigh effort51.1 %independentLiveBench
τ²-Bench
subset=banking
max effort28.3 %independentArtificial Analysis
τ²-Bench
subset=banking
high effort14.0 %independentArtificial Analysis
coding
Release 2026-06-25
tasks_counted=2 · livebench_version=2026-06-25
xhigh effort80.7 %independentLiveBench
SciCodemax effort53.6 %independentArtificial Analysis
SciCodehigh effort48.6 %independentArtificial Analysis
data analysis
Release 2026-06-25
tasks_counted=3 · livebench_version=2026-06-25
xhigh effort71.7 %independentLiveBench
external indices
AA Coding Indexmax effort71.5 pointsindependentArtificial Analysis
AA Coding Indexhigh effort66.4 pointsindependentArtificial Analysis
Intelligence Index v4.1max effort53.4 pointsindependentArtificial Analysis
Intelligence Index v4.1high effort41.7 pointsindependentArtificial Analysis
ECImax effort153.2 pointsindependentEpoch AI Benchmarking Hub
ECI153.2 pointsindependentEpoch AI Benchmarking Hub
ECIxhigh effort153.2 pointsindependentEpoch AI Benchmarking Hub
ECIhigh effort153.2 pointsindependentEpoch AI Benchmarking Hub
Vals Indexmax effort68.6 pointsindependentVals AI
factuality
SimpleQA Verifiedxhigh effort25.0 %independentEpoch AI Benchmarking Hub
2026-07-01
human preference
Coding (style control)older version
arena=text · category=coding · style_control=true
high effort1522.0
1510.1–1534.0
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
high effort1517.0
1497.4–1536.5
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
high effort1510.7
1500.7–1520.6
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
high effort1492.8
1484.7–1500.9
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
high effort1490.8
1479.8–1501.8
communityLMArena Leaderboard Dataset
Industry Life And Physical And Social Science (style control)older version
arena=text · category=industry_life_and_physical_and_social_science · style_control=true
high effort1483.2
1468.0–1498.4
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
high effort1480.8
1470.9–1490.6
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
high effort1477.8
1451.9–1503.7
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
high effort1475.7
1454.8–1496.6
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
high effort1471.0
1456.4–1485.6
communityLMArena Leaderboard Dataset
Industry Business And Management And Financial Operations (style control)older version
arena=text · category=industry_business_and_management_and_financial_operations · style_control=true
high effort1470.6
1457.1–1484.0
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
high effort1470.0
1460.8–1479.1
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
high effort1469.5
1460.6–1478.5
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
high effort1465.2
1442.3–1488.2
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
high effort1464.9
1454.2–1475.7
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
high effort1463.4
1456.4–1470.3
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
high effort1459.3
1440.6–1477.9
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
high effort1451.7
1443.1–1460.4
communityLMArena Leaderboard Dataset
Industry Writing And Literature And Language (style control)older version
arena=text · category=industry_writing_and_literature_and_language · style_control=true
high effort1448.0
1435.7–1460.2
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
high effort1430.3
1415.7–1445.0
communityLMArena Leaderboard Dataset
Industry Entertainment And Sports And Media (style control)older version
arena=text · category=industry_entertainment_and_sports_and_media · style_control=true
high effort1429.9
1417.0–1442.7
communityLMArena Leaderboard Dataset
knowledge science
GPQA Diamond
implementation=artificial-analysis
max effort91.1 %independentArtificial Analysis
GPQA Diamondxhigh effort90.5 %independentEpoch AI Benchmarking Hub
2026-07-01
GPQA Diamond
implementation=vals-ai
max effort88.9 %independentVals AI
GPQA Diamond
implementation=artificial-analysis
high effort80.0 %independentArtificial Analysis
Humanity's Last Exam
implementation=artificial-analysis
max effort39.6 %independentArtificial Analysis
Humanity's Last Exam
implementation=artificial-analysis
high effort17.8 %independentArtificial Analysis
MMLU-Pro
implementation=vals-ai
max effort87.5 %independentVals AI
language
Release 2026-06-25
tasks_counted=3 · livebench_version=2026-06-25
xhigh effort75.0 %independentLiveBench
long context instruction
AA-LCRmax effort70.7 %independentArtificial Analysis
AA-LCRhigh effort58.7 %independentArtificial Analysis
Release 2026-06-25
tasks_counted=4 · livebench_version=2026-06-25
xhigh effort63.9 %independentLiveBench
multimodal
MMMU
implementation=vals-ai
max effort83.0 %independentVals AI
professional
CorpFinmax effort67.0 %independentVals AI
LegalBenchmax effort83.9 %independentVals AI
TaxEvalmax effort75.6 %independentVals AI
reasoning math
Release 2026-06-25
tasks_counted=4 · livebench_version=2026-06-25
xhigh effort92.9 %independentLiveBench
Release 2026-06-25
tasks_counted=4 · livebench_version=2026-06-25
xhigh effort88.7 %independentLiveBench
OTIS Mock AIME 2024–2025xhigh effort94.7 %independentEpoch AI Benchmarking Hub
2026-07-01

Agent + model results

systems, not bare-model scores
agent + model Artificial Analysis harness + Claude Sonnet 5Terminal-Bench 2.180.5 %independentArtificial Analysis
agent + model Artificial Analysis harness + Claude Sonnet 5Terminal-Bench 2.175.3 %independentArtificial Analysis

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.

Related variants

Claude Sonnet 4 20250514Claude Sonnet 4.5 20250929Claude Sonnet 4.6Claude Sonnet 4.5Claude Sonnet 4