Hy3 vs Step 5 Preview
Which is better for the money: Hy3 by Tencent or Step 5 Preview by StepFun? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-25.
Trade-off
Verdict
Depends on your budget
Step 5 Preview scores 18.4 points higher (43.7 vs 25.3) but costs 8.0× as much to run the benchmark ($1190 vs $150; premium vs mid-range cost classes). They are not alternatives at one budget: pick Step 5 Preview when you need intelligence above 25.3, and Hy3 when 25.3 is enough.
| Metric | Hy3 | Step 5 Preview | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 25.3 | 43.7 | +72.7% Step 5 Preview is 72.7% more intelligent |
| Total AA cost (lower is better) | $149.61 | $1,190.26 | +695.6% Step 5 Preview is 8.0× more expensive |
| End-to-end time (lower is better) | 33.13s | 12.01s | −63.7% Step 5 Preview is 63.7% faster |
| Price per 1M tokens (in / out) | $0.14 / $0.56 | $1.00 / $2.70 | |
| Input modalities | Text | Text, Image |
Step 5 Preview vs Hy3: 72.7% more intelligence, 695.6% more cost to run the AA benchmark, and 63.7% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Hy3 is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
- Step 5 Preview is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
Hy3Step 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/hy3/step-5 · Markdown: /api/v1/compare/hy3/step-5?format=md · llms.txt · agent skill