Qwen3.8-Flash-Next vs Step 5 Preview
Which is better for the money: Qwen3.8-Flash-Next by Alibaba 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 3.9 points higher (43.7 vs 39.8) but costs 3.3× as much to run the benchmark ($1190 vs $363; premium vs mid-range cost classes). They are not alternatives at one budget: pick Step 5 Preview when you need intelligence above 39.8, and Qwen3.8-Flash-Next when 39.8 is enough.
| Metric | Qwen3.8-Flash-Next | Step 5 Preview | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 39.8 | 43.7 | +9.8% Step 5 Preview is 9.8% more intelligent |
| Total AA cost (lower is better) | $362.72 | $1,190.26 | +228.1% Step 5 Preview is 3.3× more expensive |
| End-to-end time (lower is better) | 50.57s | 12.06s | −76.2% Step 5 Preview is 76.2% faster |
| Price per 1M tokens (in / out) | $0.15 / $0.47 | $1.00 / $2.70 | |
| Input modalities | Text, Image, Video | Text, Image |
Step 5 Preview vs Qwen3.8-Flash-Next: 9.8% more intelligence, 228.1% more cost to run the AA benchmark, and 76.2% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Qwen3.8-Flash-Next is off the frontier: at the same or lower cost, GPT-6.1 Sol (Medium) scores 47.80 for $361.
- Step 5 Preview is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
Qwen3.8-Flash-NextStep 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/qwen3-8-flash-next/step-5 · Markdown: /api/v1/compare/qwen3-8-flash-next/step-5?format=md · llms.txt · agent skill