GPT-6 Luna (Max) vs Step 5 Preview
Which is better for the money: GPT-6 Luna (Max) by OpenAI or Step 5 Preview by StepFun? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-24.
Trade-off
Verdict
Depends on your budget
Step 5 Preview scores 5.6 points higher (43.7 vs 38.1) but costs 9.8× as much to run the benchmark ($1190 vs $122; premium vs mid-range cost classes). They are not alternatives at one budget: pick Step 5 Preview when you need intelligence above 38.1, and GPT-6 Luna (Max) when 38.1 is enough.
| Metric | GPT-6 Luna (Max) | Step 5 Preview | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 38.1 | 43.7 | +14.7% Step 5 Preview is 14.7% more intelligent |
| Total AA cost (lower is better) | $122.05 | $1,190.26 | +875.2% Step 5 Preview is 9.8× more expensive |
| End-to-end time (lower is better) | 82.02s | 12.06s | −85.3% Step 5 Preview is 85.3% faster |
| Price per 1M tokens (in / out) | $0.10 / $0.50 | $1.00 / $2.70 | |
| Input modalities | Text, Image | Text, Image |
Step 5 Preview vs GPT-6 Luna (Max): 14.7% more intelligence, 875.2% more cost to run the AA benchmark, and 85.3% faster end-to-end.
Today's Pareto frontier
Against the frontier
- GPT-6 Luna (Max) is on today's frontier: nothing is both smarter and cheaper.
- Step 5 Preview is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
GPT-6 Luna (Max)Step 5 PreviewToday's Pareto frontierOpen-weights frontierEvery other model
Matchup
Compare something else
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/gpt-6-luna/step-5 · Markdown: /api/v1/compare/gpt-6-luna/step-5?format=md · llms.txt · agent skill