Gemini 3.1 Pro Preview vs Gemini 3 Pro Preview
| Spec | Gemini 3.1 Pro Preview | Gemini 3 Pro Preview |
|---|---|---|
| Respan Index | 43.6 (#61) | 39.7 (#69) |
| Blended price per 1M | $4.5 | - |
| Context window | 1.0M | 1.0M |
| Input / output per 1M | $2 / $12 | - |
| Max output | 66K | 66K |
| Input | text, vision, audio | text, vision, audio, video |
| Released | 2026-02-19 | 2025-11-18 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | Gemini 3.1 Pro Preview | Gemini 3 Pro Preview |
|---|---|---|
| Coding | 14.7 Gemini 3.1 Pro Preview | 18.4 Gemini 3 Pro Preview |
| Agents | 53.4 Gemini 3.1 Pro Preview | 46.9 Gemini 3 Pro Preview |
| Reasoning | 45.8 Gemini 3.1 Pro Preview | 43.0 Gemini 3 Pro Preview |
| Knowledge & language | 72.5 Gemini 3.1 Pro Preview | 47.5 Gemini 3 Pro Preview |
| Human preference | 52.5 Gemini 3.1 Pro Preview | 52.4 Gemini 3 Pro Preview |
Benchmarks both report
Gemini 3.1 Pro Preview leads on 13 of the 16 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Gemini 3.1 Pro Preview | Gemini 3 Pro Preview |
|---|---|---|
| LMArena WebDev (LMArena) | 1446 | 1441 |
| LMArena Elo | 1487 | 1486 |
| LMArena Vision (LMArena) | 1280 | 1289 |
| Chess Puzzles (Epoch AI) | 51% | 31% |
| GPQA Diamond | 94.3% | 91.9% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 95.6% | 91.4% |
| SAGE (Vals AI) | 48.7% | 47.6% |
| SimpleQA Verified | 73.5% | 72.1% |
| SWE-bench Verified (Epoch AI) | 75.6% | 72.9% |
| Vending-Bench 2 (Andon Labs) | 3774.25 | 5478.16 |
| WeirdML (Håvard Tveit Ihle) | 72.1% | 69.9% |
| AIME 2026 | 98.3% | 91.7% |
| tau2-bench Banking Knowledge (Sierra) | 26% | 18% |
| GSO Opt@1 (GSO) | 21.6% | 17.7% |
| BALROG (BALROG) | 57% | 58.1% |
| SimpleBench (SimpleBench) | 79.6% | 76.4% |
FAQ
Is Gemini 3.1 Pro Preview better than Gemini 3 Pro Preview?
Gemini 3.1 Pro Preview scores 43.6 on the Respan Index (#61), ahead of Gemini 3 Pro Preview at 39.7 (#69). Gemini 3.1 Pro Preview leads on 13 of the 16 benchmarks both report.
Which has the larger context window?
Both accept 1.0M tokens of context.
Compare other models
- vsClaude Sonnet 5.5 vs Gemini 3.1 Pro Preview
- vsGPT-5.6 Luna vs Gemini 3.1 Pro Preview
- vsGemini 3.1 Pro Preview vs Mistral Large 4
- vsGemini 3.1 Pro Preview vs GLM-5.2
- vsGemini 3.1 Pro Preview vs Gemini 3.5 Flash
- vsGemini 3 Pro Preview vs Claude Opus 4.5
- vsGemini 3 Pro Preview vs GPT-5
- vsGemini 3 Pro Preview vs Dola Seed 2.0 Pro