Model comparison
Gemini 3 Flash Preview vs GPT-5.6 Sol
GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 52.3 on the Noometry Index. Gemini 3 Flash Preview costs 7.1× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.
Last verified . 45 shared benchmarks.
GoogleSummary
- They share 45 benchmarks with published results for both. Gemini 3 Flash Preview scores higher in 1 category and GPT-5.6 Sol in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Sol leads 85.6 to 51.7.
- The biggest single-benchmark swing is ProofBench: 15% for Gemini 3 Flash Preview and 83% for GPT-5.6 Sol.
- Gemini 3 Flash Preview is cheaper at $0.50 / $3 per million input/output tokens, against $4 / $20 for GPT-5.6 Sol.
- GPT-5.6 Sol accepts more context: 1.05M tokens versus 1.05M.
Side by side
| Gemini 3 Flash Preview | GPT-5.6 Sol | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 52.3 | 65.0 |
| Released | 2025-12-17 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.50 | $4 |
| Output $ / M tokens | $3 | $20 |
| Results tracked | 59 | 65 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Sol leads
Gemini 3 Flash Preview: 50.9 (#42), GPT-5.6 Sol: 65.1 (#7)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| LMArena WebDev | 1439 | 1618 |
| GSO | 9.8% | 76.5% |
| WeirdML | 61.6% | 89.4% |
| LMArena Coding | 1460 | 1498 |
| ALE-Bench | 1,367 | 2,177 |
| SWE-bench Verified | 75.4% | — |
| DeepSWE | — | 72.7% |
| FrontierCode | — | 47.5% |
| SWE-bench Verified (bash only) | 75.8% | — |
| CursorBench | — | 41.7% |
| SWE-bench Multilingual | 72.7% | — |
| FrontierSWE | — | 32.2% |
| SciCode | — | 57.1% |
| MirrorCode | — | 20% |
Agentic & Tool Use GPT-5.6 Sol leads
Gemini 3 Flash Preview: 38.7 (#29), GPT-5.6 Sol: 50.3 (#7)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| τ²-bench Banking | 27.3% | 46.9% |
| BALROG | 48.1% | 60% |
| GDP.pdf | 10% | 30.7% |
| LMArena Search | 1198 | 1257 |
| Vending-Bench 2 | 3,635 | 9,619 |
| Terminal-Bench | 64.3% | — |
| APEX-Agents | — | 51.4% |
| OSWorld 2.0 | — | 27.3% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Retail | 76.8% | — |
| τ²-bench Telecom | 91.2% | — |
| DeepResearch Bench | 49.8% | — |
| PostTrainBench | — | 36.2% |
| GBAEval | — | 52.6% |
Reasoning GPT-5.6 Sol leads
Gemini 3 Flash Preview: 49.2 (#37), GPT-5.6 Sol: 74.8 (#8)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| ARC-AGI-2 | 33.6% | 92.5% |
| SimpleBench | 61.1% | 71.7% |
| NYT Connections (extended) | 83.1% | 93.8% |
| ARC-AGI-1 | 84.7% | 97.5% |
| Chess Puzzles | 40% | 64% |
| LMArena Hard Prompts | 1465 | 1484 |
| Mystery Game Puzzles | 26% | 58% |
| DTBench | 89.1% | 96% |
| LMCA | 43.1% | 59.2% |
| Epoch Capabilities Index | 151.8 | 161.66 |
| Kagi LLM Benchmark | — | 67% |
| CritPt | — | 32.3% |
| EnigmaEval | — | 37.1% |
| EBR-Bench | — | 44.8% |
| Surface Evolver Bench | — | 93.1% |
| Bench to the Future 3 | — | 0.14 |
| ForecastBench | 58.5 | — |
Math GPT-5.6 Sol leads
Gemini 3 Flash Preview: 51.7 (#55), GPT-5.6 Sol: 85.6 (#9)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| FrontierMath (Tiers 1-3) | 51.2% | 89.1% |
| FrontierMath Tier 4 | 17.1% | 82.9% |
| OTIS Mock AIME 2024-2025 | 95.6% | 100% |
| ProofBench | 15% | 83% |
| LMArena Math | 1473 | 1474 |
| MathArena Final-Answer Competitions | 67.6% | — |
| FrontierMath (Feb 2025 set) | 35.6% | — |
| FrontierMath Erdős | — | 0% |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge GPT-5.6 Sol leads
Gemini 3 Flash Preview: 58.8 (#33), GPT-5.6 Sol: 64.3 (#18)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| GPQA Diamond | 89.4% | 93.5% |
| SimpleQA Verified | 66.8% | 69.7% |
| Vectara Hallucination Rate | 13.5% | 12.4% |
| LMArena Expert | 1462 | 1516 |
Multimodal GPT-5.6 Sol leads
Gemini 3 Flash Preview: 45.5 (#16), GPT-5.6 Sol: 48.6 (#9)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| LMArena Vision | 1285 | 1281 |
| Blueprint-Bench 2 | 0% | 33.6% |
| LMArena Document | 1413 | 1483 |
| GeoBench | 88% | — |
| VPCT | 72.6% | — |
| Furniture Assembly | — | 56.7% |
Multilingual Too close to call
Gemini 3 Flash Preview: 55.7 (#27), GPT-5.6 Sol: 55.3 (#32)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| LMArena Non-English | 1458 | 1452 |
| LMArena Chinese | 1511 | 1527 |
| LMArena French | 1477 | 1477 |
| LMArena German | 1497 | 1476 |
| LMArena Japanese | 1489 | 1471 |
| LMArena Korean | 1443 | 1442 |
| LMArena Russian | 1480 | 1468 |
| LMArena Spanish | 1469 | 1441 |
Instruction Following GPT-5.6 Sol leads
Gemini 3 Flash Preview: 75.7 (#56), GPT-5.6 Sol: 77.7 (#16)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| LMArena Instruction Following | 1437 | 1482 |
Long Context Too close to call
Gemini 3 Flash Preview: 44.4 (#67), GPT-5.6 Sol: 45.4 (#42)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| LMArena Longer Query | 1452 | 1480 |
Writing & Preference GPT-5.6 Sol leads
Gemini 3 Flash Preview: 65.5 (#45), GPT-5.6 Sol: 73.3 (#12)
| Benchmark | Gemini 3 Flash Preview | GPT-5.6 Sol |
|---|---|---|
| LMArena Text | 1466 | 1457 |
| LMArena Creative Writing | 1457 | 1448 |
| LMArena Multi-Turn | 1471 | 1460 |
| EQ-Bench Creative Writing | — | 1972 |
| EQ-Bench 4 | — | 1250 |
Frequently asked questions
Is Gemini 3 Flash Preview better than GPT-5.6 Sol?
GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 52.3 on the Noometry Index. Gemini 3 Flash Preview costs 7.1× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.
Which is cheaper, Gemini 3 Flash Preview or GPT-5.6 Sol?
Gemini 3 Flash Preview is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; GPT-5.6 Sol lists at $4 and $20.
Is Gemini 3 Flash Preview or GPT-5.6 Sol better for coding?
GPT-5.6 Sol scores higher on coding benchmarks: 65.1 versus 50.9 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Sol does, with 1.05M tokens against 1.05M.
How many benchmarks do Gemini 3 Flash Preview and GPT-5.6 Sol share?
45 benchmarks have published results for both models. Gemini 3 Flash Preview has 59 scored results on Noometry and GPT-5.6 Sol has 65.