The Model Index
Up to four models

Claude Sonnet 5 vs DeepSeek-V4-Pro

On the index (GPT-4 = 100), Claude Sonnet 5 scores 164 and DeepSeek-V4-Pro 163. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Claude Sonnet 5
Presets
Claude Sonnet 5
164148–180
WeightsClosed
ReleasedJun 2026
Price in / out$2.0 / $10
Context1M
Core tests taken5 of 9
DeepSeek-V4-Pro
163147–178
WeightsOpen
ReleasedAug 2026
Price in / out$0.11 / $4.44
Context1.05M
Core tests taken6 of 9

Claude Sonnet 5 and DeepSeek-V4-Pro are too close to call (2 points apart, within the uncertainty), ahead on 3 of 5 shared core tests, at 3.4× the price.

Test by testthird-party results

Claude Sonnet 5DeepSeek-V4-Pro
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)LiveBench Data AnalysisLiveBench Data AnalysisLiveBench IFLiveBench IFLiveBench LanguageLiveBench LanguageLiveBench ReasoningLiveBench ReasoningGPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProSimpleQA VerifiedSimpleQA VerifiedMATHSFrontierMath (tiers 1-3)FrontierMath (tiers 1-3)AIME-style maths (OTIS mock)AIME-style maths (OTIS mock)LiveBench MathematicsLiveBench MathematicsProofBench v1.1ProofBench v1.1Chess puzzlesChess puzzlesFrontierMath (tier 4)FrontierMath (tier 4)CODINGTerminal-Bench 4.0Terminal-Bench 4.0LiveBench Agentic CodingLiveBench Agentic CodingLiveBench CodingLiveBench CodingCode MigrationCode MigrationCyberBench v1.1CyberBench v1.1IOIIOILiveCodeBenchLiveCodeBenchProgramBenchProgramBenchSWE-benchSWE-benchVibe Code Bench 1-100Vibe Code Bench 1-100Vibe Code Bench v1.1Vibe Code Bench v1.1MirrorCodeMirrorCodeWeirdMLWeirdMLAGENTSAPEX-AgentsAPEX-AgentsSkillsBenchSkillsBenchSRE BenchSRE BenchTerminal-Bench-ScienceTerminal-Bench-ScienceDTBenchDTBenchLMCALMCAMystery game puzzlesMystery game puzzlesPROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)CorpFin v2CorpFin v2EMBEMBLegalBenchLegalBenchLegal Research BenchLegal Research BenchMedCodeMedCodeMedScribeMedScribeTax Agent BenchTax Agent BenchTaxEval v2TaxEval v2
Every result as a table
TestClaude Sonnet 5DeepSeek-V4-Pro
Reasoning & knowledge
GPQA Diamond (Epoch AI)90.5%91.7%
LiveBench Data Analysis71.7%79.2%
LiveBench IF63.9%67.7%
LiveBench Language75.0%82.1%
LiveBench Reasoning88.7%85.8%
GPQA Diamond (Vals AI)88.9%92.4%
MMLU Pro87.5%87.0%
SimpleQA Verified33.7%52.9%
Maths
FrontierMath (tiers 1-3)65.6%64.6%
AIME-style maths (OTIS mock)94.7%98.6%
LiveBench Mathematics92.9%95.1%
ProofBench v1.177.0%50.0%
Chess puzzles45.0%46.0%
FrontierMath (tier 4)29.3%26.8%
Coding
Terminal-Bench 4.010.6%14.1%
LiveBench Agentic Coding59.4%54.9%
LiveBench Coding80.7%77.2%
Code Migration44.4%41.5%
CyberBench v1.170.3%72.9%
IOI45.0%51.6%
LiveCodeBench82.4%87.5%
ProgramBench0.0%0.0%
SWE-bench79.6%96.4%
Vibe Code Bench 1-10013.8%17.5%
Vibe Code Bench v1.181.3%82.3%
MirrorCode63.3%33.3%
WeirdML69.0%66.2%
Agents
APEX-Agents54.5%47.3%
SkillsBench46.5%53.8%
SRE Bench0.4%1.9%
Terminal-Bench-Science5.7%4.3%
DTBench92.5%93.9%
LMCA50.0%45.5%
Mystery game puzzles35.0%43.0%
Professional work
Finance Agent (v2)53.9%50.4%
CorpFin v267.0%65.4%
EMB66.3%52.8%
LegalBench83.9%82.4%
Legal Research Bench41.8%40.9%
MedCode47.5%42.5%
MedScribe76.0%80.2%
Tax Agent Bench29.4%26.8%
TaxEval v275.6%73.1%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG