Gemma 4 E4B (Reasoning) vs Step 5 Preview
Which is better for the money: Gemma 4 E4B (Reasoning) by Google 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 34.8 points higher (43.7 vs 8.9) but costs 81× as much to run the benchmark ($1190 vs $14.7; premium vs budget cost classes). They are not alternatives at one budget: pick Step 5 Preview when you need intelligence above 8.9, and Gemma 4 E4B (Reasoning) when 8.9 is enough.
| Metric | Gemma 4 E4B (Reasoning) | Step 5 Preview | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 8.9 | 43.7 | +391.0% Step 5 Preview is 4.9× more intelligent |
| Total AA cost (lower is better) | $14.68 | $1,190.26 | +8008.0% Step 5 Preview is 81× more expensive |
| End-to-end time (lower is better) | 57.75s | 12.06s | −79.1% Step 5 Preview is 79.1% faster |
| Price per 1M tokens (in / out) | $0.02 / $0.10 | $1.00 / $2.70 | |
| Input modalities | Text, Image, Speech, Video | Text, Image |
Step 5 Preview vs Gemma 4 E4B (Reasoning): 391.0% more intelligence, 8008.0% more cost to run the AA benchmark, and 79.1% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Gemma 4 E4B (Reasoning) is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
- Step 5 Preview is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
Gemma 4 E4B (Reasoning)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/gemma-4-e4b/step-5 · Markdown: /api/v1/compare/gemma-4-e4b/step-5?format=md · llms.txt · agent skill