The Model Index
Up to four models

Claude Opus 5.5 vs Hy4 preview

On the index (GPT-4 = 100), Claude Opus 5.5 scores 195 and Hy4 preview 173. 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 Opus 5.5
Presets
Claude Opus 5.5
195178–211
WeightsClosed
ReleasedSep 2026
Price in / out$4.0 / $20
Context1M
Core tests taken6 of 9
Hy4 preview
173154–192
WeightsOpen
ReleasedAug 2026
Price in / out$0.83 / $2.5
Context1.05M
Core tests taken2 of 9

Claude Opus 5.5 leads Hy4 preview by 22 points, ahead on 2 of 2 shared core tests, at 6.4× the price.

Test by testthird-party results

Claude Opus 5.5Hy4 preview
0%25%50%75%100%MATHSProofBench v1.1ProofBench v1.1CODINGTerminal-Bench 4.0Terminal-Bench 4.0Code MigrationCode MigrationCyberBench v1.1CyberBench v1.1IOIIOIProgramBenchProgramBenchVibe Code Bench v1.1Vibe Code Bench v1.1AGENTSBioMysteryBenchBioMysteryBenchSRE BenchSRE BenchPROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)EMBEMBHLAB-Vals-1.1HLAB-Vals-1.1Legal Research BenchLegal Research BenchMedCodeMedCodeMedScribeMedScribePublic Benefits Bench v1.1Public Benefits Bench v1.1Tax Agent BenchTax Agent Bench
Every result as a table
TestClaude Opus 5.5Hy4 preview
Maths
ProofBench v1.1100.0%75.0%
Coding
Terminal-Bench 4.065.1%39.4%
Code Migration66.6%47.4%
CyberBench v1.174.6%65.4%
IOI95.1%59.3%
ProgramBench18.5%0.0%
Vibe Code Bench v1.190.3%81.0%
Agents
BioMysteryBench79.3%69.3%
SRE Bench33.6%2.3%
Professional work
Finance Agent (v2)58.6%55.1%
EMB75.9%58.1%
HLAB-Vals-1.19.2%12.5%
Legal Research Bench50.5%45.2%
MedCode49.8%43.3%
MedScribe91.4%83.6%
Public Benefits Bench v1.170.6%68.6%
Tax Agent Bench44.5%34.4%

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

This page as a chartDownload PNG