Gemini 3.1 Pro Preview vs Gemini 3.5 Flash (High)
Which is better for the money: Gemini 3.1 Pro Preview by Google or Gemini 3.5 Flash (High) by Google? 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
Gemini 3.5 Flash (High) scores 2.9 points higher (32.6 vs 29.7) but costs 2.2× as much to run the benchmark ($4222 vs $1935; flagship vs premium cost classes). Pick Gemini 3.5 Flash (High) when the extra intelligence is worth the premium, otherwise Gemini 3.1 Pro Preview.
| Metric | Gemini 3.1 Pro Preview | Gemini 3.5 Flash (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 29.7 | 32.6 | +9.8% Gemini 3.5 Flash (Hig… is 9.8% more intelligent |
| Total AA cost (lower is better) | $1,935.39 | $4,222.12 | +118.2% Gemini 3.5 Flash (Hig… is 2.2× more expensive |
| End-to-end time (lower is better) | 24.47s | 16.63s | −32.0% Gemini 3.5 Flash (Hig… is 32.0% faster |
| Price per 1M tokens (in / out) | $2.00 / $12.0 | $1.50 / $9.00 | |
| Input modalities | Text, Image, Speech, Video | Text, Image, Speech, Video |
Gemini 3.5 Flash (High) vs Gemini 3.1 Pro Preview: 9.8% more intelligence, 118.2% more cost to run the AA benchmark, and 32.0% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Gemini 3.1 Pro Preview is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
- Gemini 3.5 Flash (High) is off the frontier: at the same or lower cost, Claude Opus 5.5 (Xhigh, Default Fallback) scores 56.00 for $4057.
Gemini 3.1 Pro PreviewGemini 3.5 Flash (High)Today'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/gemini-3-1-pro-preview/gemini-3-5-flash · Markdown: /api/v1/compare/gemini-3-1-pro-preview/gemini-3-5-flash?format=md · llms.txt · agent skill