Celeris-1 vs Step 5 Preview
Which is better for the money: Celeris-1 by Celeris or Step 5 Preview by StepFun? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-09-22.
Trade-off
Verdict
Depends on your budget
Step 5 Preview scores 37.4 points higher (43.7 vs 6.3) but costs 15× as much to run the benchmark ($925 vs $60.0; premium vs mid-range cost classes). They are not alternatives at one budget: pick Step 5 Preview when you need intelligence above 6.3, and Celeris-1 when 6.3 is enough.
| Metric | Celeris-1 | Step 5 Preview | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 6.3 | 43.7 | +593.7% Step 5 Preview is 6.9× more intelligent |
| Total AA cost (lower is better) | $59.99 | $924.68 | +1441.4% Step 5 Preview is 15× more expensive |
| End-to-end time (lower is better) | 0.92s | 33.68s | +3560.9% Step 5 Preview is 3560.9% slower |
| Price per 1M tokens (in / out) | $0.20 / $0.70 | $1.00 / $2.70 | |
| Input modalities | Text, Image | Text, Image |
Step 5 Preview vs Celeris-1: 593.7% more intelligence, 1441.4% more cost to run the AA benchmark, and 3560.9% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Celeris-1 is off the frontier: at the same or lower cost, GPT-5.6 Luna (medium) scores 25.00 for $44.3.
- Step 5 Preview is off the frontier: at the same or lower cost, MiMo-V2.6-Pro scores 46.30 for $207.
Celeris-1Step 5 PreviewToday's Pareto frontierEvery other model
Matchup
Compare something else
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/celeris-1/step-5 · Markdown: /api/v1/compare/celeris-1/step-5?format=md · llms.txt · agent skill