The Model Index
Up to four models

Hy4 preview vs Grok 4.6

On the index (GPT-4 = 100), Hy4 preview scores 173 and Grok 4.6 167. 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 Hy4 preview
Presets
Hy4 preview
173154–192
WeightsOpen
ReleasedAug 2026
Price in / out$0.83 / $2.5
Context1.05M
Core tests taken2 of 9
Grok 4.6
167151–183
WeightsClosed
ReleasedAug 2026
Price in / out$2.0 / $6.0
Context500K
Core tests taken6 of 9

Hy4 preview and Grok 4.6 are too close to call (6 points apart, within the uncertainty), ahead on 2 of 2 shared core tests, and costs 2.4× less.

Test by testthird-party results

Hy4 previewGrok 4.6
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 v1.1AGENTSBioMysteryBenchBioMysteryBenchPROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)EMBEMBLegalBenchLegalBenchLegal Research BenchLegal Research BenchMedCodeMedCodeMedScribeMedScribePublic Benefits Bench v1.1Public Benefits Bench v1.1Tax Agent BenchTax Agent Bench
Every result as a table
TestHy4 previewGrok 4.6
Maths
ProofBench v1.175.0%51.0%
Coding
Terminal-Bench 4.039.4%17.2%
Code Migration47.4%44.6%
CyberBench v1.165.4%67.7%
IOI59.3%47.6%
ProgramBench0.0%1.0%
Vibe Code Bench v1.181.0%76.2%
Agents
BioMysteryBench69.3%72.2%
Professional work
Finance Agent (v2)55.1%53.7%
EMB58.1%62.7%
LegalBench83.8%86.3%
Legal Research Bench45.2%48.1%
MedCode43.3%44.7%
MedScribe83.6%86.5%
Public Benefits Bench v1.168.6%66.8%
Tax Agent Bench34.4%33.7%

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

This page as a chartDownload PNG