Compare / head-to-head
Gemini 3 Pro Previewvs
GPT-6 Sol
Gemini 3 Pro Preview and GPT-6 Sol split their 6 shared benchmarks evenly. GPT-6 Sol is 1.2x cheaper per token. GPT-6 Sol has the larger context window (1.1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
3 – 3
even split
Cheaper per token
GPT-6 Sol
1.2x cheaper, input + output
Larger context
GPT-6 Sol
1.1M tokens
Providers
2 providers
Google · OpenAI
Gemini 3 Pro Preview
Google · released 2025-11-18
textvisionaudiovideo
Quality
Benchmark matrix
6 shared · 18 only Gemini 3 Pro Preview · 21 only GPT-6 Sol| Benchmark | Gemini 3 Pro Preview | GPT-6 Sol | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 6 | ||||
| GPQA Diamond | 91.9% ↗ | 94.3% ↗epoch run | -2.4 pt | GPT-6 Sol |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 91.4% ↗epoch run | 100% ↗epoch run | -8.6 pt | GPT-6 Sol |
| SAGE (Vals AI) | 47.6% ↗ | 44.8% ↗ | +2.8 pt | Gemini 3 Pro Preview |
| SimpleBench (SimpleBench) | 76.4% ↗ | 73.1% ↗ | +3.3 pt | Gemini 3 Pro Preview |
| SimpleQA Verified | 72.1% ↗ | 60.7% ↗epoch run | +11.4 pt | Gemini 3 Pro Preview |
| Vending-Bench 2 (Andon Labs) | 5478.16 ↗ | 14427.85 ↗ | -8949.7 | GPT-6 Sol |
| Only Gemini 3 Pro Preview reports · 18 | ||||
| SWE-bench Verified | 76.2% ↗ | not reported | — | — |
| AIME 2025 | 95% ↗ | not reported | — | — |
| Humanity's Last Exam (no tools) | 37.5% ↗ | not reported | — | — |
| AIME 2025 (with code execution) | 100% ↗ | not reported | — | — |
| ARC-AGI-2 (ARC Prize Verified) | 31.1% ↗ | not reported | — | — |
| BALROG (BALROG) | 58.1% ↗ | not reported | — | — |
| CharXiv Reasoning | 81.4% ↗ | not reported | — | — |
| Chess Puzzles (Epoch AI run) | 31% ↗epoch run | not reported | — | — |
| GSO Opt@1 (GSO) | 17.7% ↗ | not reported | — | — |
| Humanity's Last Exam (with search and code execution) | 45.8% ↗ | not reported | — | — |
| MathArena Apex | 23.4% ↗ | not reported | — | — |
| MMMLU | 91.8% ↗ | not reported | — | — |
| MMMU-Pro | 81% ↗ | not reported | — | — |
| SWE-bench Verified (Epoch AI run) | 72.9% ↗epoch run | not reported | — | — |
| Terminal-Bench 2.0 (Terminus-2 agent) | 54.2% ↗ | not reported | — | — |
| Video-MMMU | 87.6% ↗ | not reported | — | — |
| WeirdML (Håvard Tveit Ihle) | 69.9% ↗ | not reported | — | — |
| τ2-bench | 85.4% ↗ | not reported | — | — |
| Only GPT-6 Sol reports · 21 | ||||
| LMArena Elo | not reported | 1457.2 ↗ | — | — |
| Agents' Last Exam V1 | not reported | 56.4% ↗ | — | — |
| APEX-Agents (Mercor) | not reported | 54.3% ↗ | — | — |
| AutomationBench 1.0.6 | not reported | 33.2% ↗ | — | — |
| DeepSWE v1.1 | not reported | 68.8% ↗ | — | — |
| EBR-bench (Epoch AI run) | not reported | 53.3% ↗epoch run | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | not reported | 90% ↗epoch run | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | not reported | 89.8% ↗epoch run | — | — |
| Furniture Assembly (Epoch AI run) | not reported | 58.3% ↗epoch run | — | — |
| LiveBench Agentic Coding (LiveBench) | not reported | 52.9% ↗ | — | — |
| LiveBench Coding (LiveBench) | not reported | 81.8% ↗ | — | — |
| LiveBench Data Analysis (LiveBench) | not reported | 81.2% ↗ | — | — |
| LiveBench Instruction Following (LiveBench) | not reported | 68.6% ↗ | — | — |
| LiveBench Language (LiveBench) | not reported | 85.3% ↗ | — | — |
| LiveBench Mathematics (LiveBench) | not reported | 96.4% ↗ | — | — |
| LiveBench Reasoning (LiveBench) | not reported | 88.7% ↗ | — | — |
| LMArena Agent (LMArena) | not reported | 0.0971 ↗ | — | — |
| LMArena WebDev (LMArena) | not reported | 1689 ↗ | — | — |
| Mystery Game Puzzles (Epoch AI run) | not reported | 56% ↗epoch run | — | — |
| OSWorld 2.0 (offline, partial) | not reported | 60.5% ↗ | — | — |
| Terminal-Bench 4.0 (Vals AI) | not reported | 44.4% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Gemini 3 Pro Preview minus GPT-6 Sol in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | Gemini 3 Pro Preview | GPT-6 Sol | Edge |
|---|---|---|---|
| Context window | 1.0M tokens | 1.1M tokens | GPT-6 Sol |
| Max output | 66K tokens | 128K tokens | GPT-6 Sol |
| Input price / 1M | $2 ↗ | $2 ↗ | Tie |
| Output price / 1M | $12 ↗ | $10 ↗ | GPT-6 Sol |
| Cached input / 1M | — | $0.2 ↗ | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $14 | $12 | GPT-6 Sol |
| Modalities | text · vision · audio · video | text · vision | Gemini 3 Pro Preview |
| Released | 2025-11-18 | 2026-09-22 | — |
| Cited benchmark scores | 25 | 27 | — |
Reliability
Provider status
Live from /statusMore matchups
Gemini 3 Pro Preview vs …
Models sharing the most benchmarksMore matchups
GPT-6 Sol vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Gemini 3 Pro Preview and GPT-6 Sol on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.