Gemini 3.1 Pro Preview vs GPT-6 Astra (Medium)
Which is better for the money: Gemini 3.1 Pro Preview by Google or GPT-6 Astra (Medium) by OpenAI? 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
GPT-6 Astra (Medium) scores 19.9 points higher (49.6 vs 29.7) but costs 1.3× as much to run the benchmark ($2434 vs $1935; flagship vs premium cost classes). Pick GPT-6 Astra (Medium) when the extra intelligence is worth the premium, otherwise Gemini 3.1 Pro Preview.
| Metric | Gemini 3.1 Pro Preview | GPT-6 Astra (Medium) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 29.7 | 49.6 | +67.0% GPT-6 Astra (Medium) is 67.0% more intelligent |
| Total AA cost (lower is better) | $1,935.39 | $2,434.12 | +25.8% GPT-6 Astra (Medium) is 25.8% more expensive |
| End-to-end time (lower is better) | 25.04s | 13.89s | −44.5% GPT-6 Astra (Medium) is 44.5% faster |
| Price per 1M tokens (in / out) | $2.00 / $12.0 | $10.0 / $50.0 | |
| Input modalities | Text, Image, Speech, Video | Text, Image |
GPT-6 Astra (Medium) vs Gemini 3.1 Pro Preview: 67.0% more intelligence, 25.8% more cost to run the AA benchmark, and 44.5% 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.
- GPT-6 Astra (Medium) is off the frontier: at the same or lower cost, Claude Opus 5.5 (High, Default Fallback) scores 53.60 for $2172.
Gemini 3.1 Pro PreviewGPT-6 Astra (Medium)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/gpt-6-astra-medium · Markdown: /api/v1/compare/gemini-3-1-pro-preview/gpt-6-astra-medium?format=md · llms.txt · agent skill