# Gemini 3.7 Flash vs GPT-4o

> Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 28.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-7-flash-vs-gpt-4o
- Last updated: 2026-10-10
- Shared benchmarks: 29

## Summary

- They share 29 benchmarks with published results for both. Gemini 3.7 Flash scores higher in 10 categories and GPT-4o in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.7 Flash leads 70.0 to 9.4.
- The biggest single-benchmark swing is ARC-AGI-1: 95.5% for Gemini 3.7 Flash and 4.5% for GPT-4o.
- Gemini 3.7 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $2.50 / $10 for GPT-4o.
- Gemini 3.7 Flash accepts more context: 1.05M tokens versus 128K.

## Snapshot

| | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 59.8 | 28.6 |
| Rank | 14 | 324 |
| Context | 1.05M | 128K |
| Input $/M | $0.75 | $2.50 |
| Output $/M | $3.75 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.7 Flash: 56.2 (#22)
- GPT-4o: 24.8 (#328)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| LMArena Coding | 1497 | 1297 |
| SWE-bench Verified | — | 31% |
| DeepSWE | 65.5% | — |
| FrontierCode | 43.6% | — |
| SWE-bench Verified (bash only) | — | 21.6% |
| Aider Polyglot | — | 45.3% |
| LMArena WebDev | 1592 | — |
| FrontierSWE | 20.3% | — |
| SciCode | 59.8% | — |
| GSO | — | 0% |
| WeirdML | — | 25.1% |
| BigCodeBench Instruct | — | 51.1% |
| LiveBench Coding | — | 51.4% |
| BigCodeBench Complete | — | 61.1% |
| CadEval | — | 26% |
| ALE-Bench | 904.3 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |

## Agentic & Tool Use

- Gemini 3.7 Flash: 42.1 (#19)
- GPT-4o: 21.0 (#141)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| APEX-Agents | 67.8% | — |
| GDPval | — | 9.9% |
| Remote Labor Index | 5% | — |
| TheAgentCompany | — | 8.6% |
| Cybench | — | 12.5% |
| BALROG | — | 32.3% |
| GDP.pdf | 23.8% | — |
| LMArena Search | — | 1006 |
| METR Time Horizons | — | 40.8% |

## Reasoning

- Gemini 3.7 Flash: 70.0 (#15)
- GPT-4o: 9.4 (#343)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 84.6% | 0% |
| ARC-AGI-1 | 95.5% | 4.5% |
| CritPt | 14.3% | 0% |
| Chess Puzzles | 47% | 13% |
| LMArena Hard Prompts | 1494 | 1281 |
| DTBench | 96.8% | 64.5% |
| LMCA | 50.4% | 16.6% |
| Epoch Capabilities Index | 157.27 | 128.97 |
| SimpleBench | — | 17.8% |
| NYT Connections (extended) | 94% | — |
| EnigmaEval | — | 0.8% |
| LiveBench Reasoning | — | 55.8% |
| Mystery Game Puzzles | 37% | — |
| LiveBench Data Analysis | — | 60.9% |
| ForecastBench | — | 57.7 |
| LiveBench | — | 55.3% |

## Math

- Gemini 3.7 Flash: 69.6 (#23)
- GPT-4o: 10.6 (#312)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| FrontierMath (Tiers 1-3) | 71.6% | 0.4% |
| OTIS Mock AIME 2024-2025 | 97.2% | 6.4% |
| LMArena Math | 1507 | 1285 |
| FrontierMath Tier 4 | 36.6% | — |
| ProofBench | 58% | — |
| Omni-MATH | — | 29.3% |
| LiveBench Math | — | 49.5% |
| MATH Level 5 | — | 53.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |

## Knowledge

- Gemini 3.7 Flash: 69.7 (#5)
- GPT-4o: 28.8 (#242)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| GPQA Diamond | 94.8% | 49.2% |
| SimpleQA Verified | 69.2% | 26% |
| LMArena Expert | 1508 | 1250 |
| Humanity's Last Exam | — | 2.7% |
| MMLU-Pro | — | 71.3% |
| Confabulations | — | 15.3% |
| Vectara Hallucination Rate | — | 9.6% |
| GPQA (HELM) | — | 52% |
| MMLU | — | 88.1% |

## Multimodal

- Gemini 3.7 Flash: 37.3 (#73)
- GPT-4o: 34.5 (#91)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| LMArena Vision | 1316 | 1137 |
| Video-MME | — | 71.9% |
| GeoBench | — | 71% |
| VPCT | — | 40% |
| Furniture Assembly | 26.7% | — |
| ScienceQA | — | 88.5% |

## Multilingual

- Gemini 3.7 Flash: 57.6 (#7)
- GPT-4o: 43.2 (#186)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| LMArena Non-English | 1484 | 1283 |
| LMArena Chinese | 1548 | 1277 |
| LMArena French | 1505 | 1304 |
| LMArena German | 1498 | 1282 |
| LMArena Japanese | 1512 | 1257 |
| LMArena Korean | 1483 | 1234 |
| LMArena Russian | 1516 | 1286 |
| LMArena Spanish | 1503 | 1292 |

## Instruction Following

- Gemini 3.7 Flash: 77.7 (#15)
- GPT-4o: 66.6 (#207)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| LMArena Instruction Following | 1483 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
| IFEval | — | 81.7% |

## Long Context

- Gemini 3.7 Flash: 45.7 (#30)
- GPT-4o: 39.4 (#179)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| LMArena Longer Query | 1492 | 1289 |
| Fiction.LiveBench | — | 66.7% |

## Writing & Preference

- Gemini 3.7 Flash: 71.2 (#20)
- GPT-4o: 52.6 (#166)

| Benchmark | Gemini 3.7 Flash | GPT-4o |
|---|---|---|
| LMArena Text | 1486 | 1300 |
| LMArena Creative Writing | 1490 | 1292 |
| LMArena Multi-Turn | 1489 | 1302 |
| Short-Story Creative Writing | — | 81.8% |
| EQ-Bench Creative Writing | 1723 | — |
| WildBench | — | 82.8% |
| LiveBench Language | — | 47.6% |

## FAQ

### Is Gemini 3.7 Flash better than GPT-4o?

Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 28.6 on the Noometry Index.

### Which is cheaper, Gemini 3.7 Flash or GPT-4o?

Gemini 3.7 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; GPT-4o lists at $2.50 and $10.

### Is Gemini 3.7 Flash or GPT-4o better for coding?

Gemini 3.7 Flash scores higher on coding benchmarks: 56.2 versus 24.8 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.7 Flash does, with 1.05M tokens against 128K.

### How many benchmarks do Gemini 3.7 Flash and GPT-4o share?

29 benchmarks have published results for both models. Gemini 3.7 Flash has 44 scored results on Noometry and GPT-4o has 72.