Mercury 2 vs Step 5 Preview
Which is better for the money: Mercury 2 by Inception 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 32.9 points higher (43.7 vs 10.8) but costs 11× as much to run the benchmark ($1190 vs $105; premium vs mid-range cost classes). They are not alternatives at one budget: pick Step 5 Preview when you need intelligence above 10.8, and Mercury 2 when 10.8 is enough.
| Metric | Mercury 2 | Step 5 Preview | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 10.8 | 43.7 | +304.6% Step 5 Preview is 4.0× more intelligent |
| Total AA cost (lower is better) | $104.94 | $1,190.26 | +1034.2% Step 5 Preview is 11× more expensive |
| End-to-end time (lower is better) | 8.08s | 12.06s | +49.3% Step 5 Preview is 49.3% slower |
| Price per 1M tokens (in / out) | $0.25 / $0.75 | $1.00 / $2.70 | |
| Input modalities | Text | Text, Image |
Step 5 Preview vs Mercury 2: 304.6% more intelligence, 1034.2% more cost to run the AA benchmark, and 49.3% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Mercury 2 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (High) scores 37.80 for $94.1.
- Step 5 Preview is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
Mercury 2Step 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/mercury-2/step-5 · Markdown: /api/v1/compare/mercury-2/step-5?format=md · llms.txt · agent skill