Gemini 3.1 Pro Preview vs GPT-6.1 Sol
| Spec | Gemini 3.1 Pro Preview | GPT-6.1 Sol |
|---|---|---|
| Respan Index | 43.9 (#51) | 69.8 (#7) |
| Blended price per 1M | $2.25 | $2 |
| Context window | 1.0M | 1.1M |
| Input / output per 1M | $1 / $6 | $1 / $5 |
| Max output | 66K | 128K |
| Input | text, vision, audio | text, vision |
| Released | 2026-02-19 | 2026-09-29 |
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 | GPT-6.1 Sol |
|---|---|---|
| Coding | 15.6 Gemini 3.1 Pro Preview | 53.3 GPT-6.1 Sol |
| Agents | 53.6 Gemini 3.1 Pro Preview | 73.1 GPT-6.1 Sol |
| Reasoning | 46.2 Gemini 3.1 Pro Preview | 87.2 GPT-6.1 Sol |
| Knowledge & language | 72.5 Gemini 3.1 Pro Preview | 74.4 GPT-6.1 Sol |
| Human preference | 52.2 Gemini 3.1 Pro Preview | 59.8 GPT-6.1 Sol |
Benchmarks both report
GPT-6.1 Sol leads on 23 of the 26 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Gemini 3.1 Pro Preview | GPT-6.1 Sol |
|---|---|---|
| SAGE 1.1 (Vals AI) | 83.2% | 90.5% |
| APEX-Agents (Mercor) | 35.1% | 60% |
| Chess Puzzles (Epoch AI) | 51% | 68% |
| LMArena Agent (LMArena) | -0.0445 | 0.1247 |
| LMArena Elo | 1487 | 1481 |
| LMArena WebDev (LMArena) | 1447 | 1751 |
| LMArena Vision (LMArena) | 1280 | 1294 |
| EBR-bench (Epoch AI) | 14.3% | 54.3% |
| FrontierMath Tier 4 v2 (Epoch AI) | 26.8% | 100% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 59.6% | 93.7% |
| Furniture Assembly (Epoch AI) | 26.7% | 80% |
| GPQA Diamond | 94.3% | 95.4% |
| LiveBench Agentic Coding (LiveBench) | 44.1% | 56.8% |
| LiveBench Coding (LiveBench) | 76.5% | 80.7% |
| LiveBench Data Analysis (LiveBench) | 78.5% | 82.7% |
| LiveBench Instruction Following (LiveBench) | 79.1% | 74.2% |
| LiveBench Language (LiveBench) | 85.4% | 90.1% |
| LiveBench Mathematics (LiveBench) | 91% | 96.8% |
| LiveBench Reasoning (LiveBench) | 84% | 92.6% |
| MirrorCode (Epoch AI) | 8.9% | 61.1% |
| Mystery Game Puzzles (Epoch AI) | 34% | 80% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 95.6% | 100% |
| SAGE (Vals AI) | 48.7% | 46.5% |
| SimpleQA Verified | 73.5% | 73.9% |
| Terminal-Bench 4.0 (Vals AI) | 2.5% | 55.1% |
| SimpleBench (SimpleBench) | 79.6% | 82.9% |
FAQ
Is Gemini 3.1 Pro Preview better than GPT-6.1 Sol?
GPT-6.1 Sol scores 69.8 on the Respan Index (#7), ahead of Gemini 3.1 Pro Preview at 43.9 (#51). GPT-6.1 Sol leads on 23 of the 26 benchmarks both report.
Which is cheaper, Gemini 3.1 Pro Preview or GPT-6.1 Sol?
Gemini 3.1 Pro Preview lists at $1 input and $6 output per 1M tokens; GPT-6.1 Sol at $1 and $5. GPT-6.1 Sol is 1.1x cheaper per token at list price.
Which has the larger context window?
Gemini 3.1 Pro Preview accepts 1.0M tokens and GPT-6.1 Sol 1.1M tokens.