# Gemini 3.7 Flash vs GPT-6 Sol

> GPT-6 Sol is the stronger model overall, scoring 61.8 to 59.8 on the Noometry Index. Gemini 3.7 Flash costs 2.7× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemini-3-7-flash-vs-gpt-6-sol
- Last updated: 2026-10-11
- Shared benchmarks: 41

## Summary

- They share 41 benchmarks with published results for both. Gemini 3.7 Flash scores higher in 5 categories and GPT-6 Sol in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Sol leads 87.2 to 69.6.
- The biggest single-benchmark swing is FrontierMath Tier 4: 36.6% for Gemini 3.7 Flash and 90% for GPT-6 Sol.
- Gemini 3.7 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 1.05M.

## Snapshot

| | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 59.8 | 61.8 |
| Rank | 14 | 12 |
| Context | 1.05M | 1.05M |
| Input $/M | $0.75 | $2 |
| Output $/M | $3.75 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.7 Flash: 56.2 (#22)
- GPT-6 Sol: 60.1 (#11)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| DeepSWE | 65.5% | 68.8% |
| FrontierCode | 43.6% | 49.3% |
| LMArena WebDev | 1592 | 1688 |
| SciCode | 59.8% | 57.6% |
| LMArena Coding | 1497 | 1447 |
| ALE-Bench | 904.3 | 2,462 |
| FrontierSWE | 20.3% | — |

## Agentic & Tool Use

- Gemini 3.7 Flash: 42.1 (#19)
- GPT-6 Sol: 37.2 (#36)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| APEX-Agents | 67.8% | 54.3% |
| GDP.pdf | 23.8% | 26.4% |
| Remote Labor Index | 5% | — |
| Vending-Bench 2 | — | 14,428 |

## Reasoning

- Gemini 3.7 Flash: 70.0 (#15)
- GPT-6 Sol: 74.0 (#9)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| ARC-AGI-2 | 84.6% | 89.6% |
| NYT Connections (extended) | 94% | 90.1% |
| ARC-AGI-1 | 95.5% | 95.5% |
| CritPt | 14.3% | 30.9% |
| LMArena Hard Prompts | 1494 | 1418 |
| Mystery Game Puzzles | 37% | 56% |
| DTBench | 96.8% | 97.3% |
| LMCA | 50.4% | 59.1% |
| Epoch Capabilities Index | 157.27 | 162.72 |
| Chess Puzzles | 47% | — |
| EBR-Bench | — | 53.3% |

## Math

- Gemini 3.7 Flash: 69.6 (#23)
- GPT-6 Sol: 87.2 (#7)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| FrontierMath (Tiers 1-3) | 71.6% | 89.8% |
| FrontierMath Tier 4 | 36.6% | 90% |
| OTIS Mock AIME 2024-2025 | 97.2% | 100% |
| ProofBench | 58% | 83% |
| LMArena Math | 1507 | 1402 |

## Knowledge

- Gemini 3.7 Flash: 69.7 (#5)
- GPT-6 Sol: 64.8 (#15)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| GPQA Diamond | 94.8% | 94.3% |
| SimpleQA Verified | 69.2% | 60.7% |
| LMArena Expert | 1508 | 1439 |
| Vectara Hallucination Rate | — | 6.5% |

## Multimodal

- Gemini 3.7 Flash: 37.3 (#73)
- GPT-6 Sol: 47.6 (#10)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| LMArena Vision | 1316 | 1245 |
| Furniture Assembly | 26.7% | 58.3% |
| Blueprint-Bench 2 | — | 36.9% |

## Multilingual

- Gemini 3.7 Flash: 57.6 (#7)
- GPT-6 Sol: 50.5 (#118)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| LMArena Non-English | 1484 | 1385 |
| LMArena Chinese | 1548 | 1405 |
| LMArena French | 1505 | 1410 |
| LMArena German | 1498 | 1390 |
| LMArena Japanese | 1512 | 1385 |
| LMArena Korean | 1483 | 1341 |
| LMArena Russian | 1516 | 1401 |
| LMArena Spanish | 1503 | 1384 |

## Instruction Following

- Gemini 3.7 Flash: 77.7 (#15)
- GPT-6 Sol: 74.5 (#94)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| LMArena Instruction Following | 1483 | 1412 |

## Long Context

- Gemini 3.7 Flash: 45.7 (#30)
- GPT-6 Sol: 43.1 (#108)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| LMArena Longer Query | 1492 | 1411 |

## Writing & Preference

- Gemini 3.7 Flash: 71.2 (#20)
- GPT-6 Sol: 71.9 (#18)

| Benchmark | Gemini 3.7 Flash | GPT-6 Sol |
|---|---|---|
| LMArena Text | 1486 | 1395 |
| LMArena Creative Writing | 1490 | 1378 |
| EQ-Bench Creative Writing | 1723 | 2125 |
| LMArena Multi-Turn | 1489 | 1412 |

## FAQ

### Is Gemini 3.7 Flash better than GPT-6 Sol?

GPT-6 Sol is the stronger model overall, scoring 61.8 to 59.8 on the Noometry Index. Gemini 3.7 Flash costs 2.7× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.

### Which is cheaper, Gemini 3.7 Flash or GPT-6 Sol?

Gemini 3.7 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; GPT-6 Sol lists at $2 and $10.

### Is Gemini 3.7 Flash or GPT-6 Sol better for coding?

GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 56.2 in the Noometry coding category.

### Which has the bigger context window?

GPT-6 Sol does, with 1.05M tokens against 1.05M.

### How many benchmarks do Gemini 3.7 Flash and GPT-6 Sol share?

41 benchmarks have published results for both models. Gemini 3.7 Flash has 44 scored results on Noometry and GPT-6 Sol has 45.