The Model Index
Up to four models

Step 5 Preview vs GPT-6 Luna

On the index (GPT-4 = 100), Step 5 Preview scores 163 and GPT-6 Luna 162. 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 Step 5 Preview
Presets
Step 5 Preview
163145–180
WeightsClosed
ReleasedSep 2026
Price in / out$1.0 / $2.7
Context1M
Core tests taken2 of 9
GPT-6 Luna
162146–178
WeightsClosed
ReleasedSep 2026
Price in / out$0.10 / $0.50
Context1.05M
Core tests taken6 of 9

Step 5 Preview and GPT-6 Luna are too close to call (0 points apart, within the uncertainty), ahead on 2 of 2 shared core tests, at 7.1× the price.

Test by testthird-party results

Step 5 PreviewGPT-6 Luna
0%25%50%75%100%MATHSProofBench v1.1ProofBench v1.1CODINGTerminal-Bench 4.0Terminal-Bench 4.0Code MigrationCode MigrationIOIIOIVibe Code Bench v1.1Vibe Code Bench v1.1AGENTSBioMysteryBenchBioMysteryBenchSRE BenchSRE BenchTerminal-Bench-ScienceTerminal-Bench-SciencePROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)EMBEMBHLAB-Vals-1.1HLAB-Vals-1.1Legal Research BenchLegal Research BenchPublic Benefits Bench v1.1Public Benefits Bench v1.1Tax Agent BenchTax Agent BenchNOVEL PROBLEMSMysteryMechanismMysteryMechanism
Every result as a table
TestStep 5 PreviewGPT-6 Luna
Maths
ProofBench v1.142.0%64.0%
Coding
Terminal-Bench 4.031.8%13.6%
Code Migration39.0%42.5%
IOI38.1%55.6%
Vibe Code Bench v1.170.4%81.7%
Agents
BioMysteryBench67.4%61.5%
SRE Bench0.4%2.7%
Terminal-Bench-Science2.9%4.3%
Professional work
Finance Agent (v2)50.7%49.9%
EMB57.1%68.5%
HLAB-Vals-1.112.9%4.2%
Legal Research Bench40.4%30.3%
Public Benefits Bench v1.158.1%57.6%
Tax Agent Bench27.3%20.0%
Novel problems
MysteryMechanism18.5%19.4%

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

This page as a chartDownload PNG