# Gemini 3.7 Flash vs o4-mini

> Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 41.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-7-flash-vs-o4-mini
- Last updated: 2026-10-11
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. Gemini 3.7 Flash scores higher in 9 categories and o4-mini in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.7 Flash leads 70.0 to 24.6.
- The biggest single-benchmark swing is ARC-AGI-2: 84.6% for Gemini 3.7 Flash and 6.1% for o4-mini.
- Gemini 3.7 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $1.10 / $4.40 for o4-mini.
- Gemini 3.7 Flash accepts more context: 1.05M tokens versus 200K.

## Snapshot

| | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 59.8 | 41.6 |
| Rank | 14 | 132 |
| Context | 1.05M | 200K |
| Input $/M | $0.75 | $1.10 |
| Output $/M | $3.75 | $4.40 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.7 Flash: 56.2 (#22)
- o4-mini: 40.9 (#127)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| LMArena Coding | 1497 | 1368 |
| ALE-Bench | 904.3 | 826.17 |
| DeepSWE | 65.5% | — |
| FrontierCode | 43.6% | — |
| SWE-bench Verified (bash only) | — | 45% |
| Aider Polyglot | — | 72% |
| LMArena WebDev | 1592 | — |
| FrontierSWE | 20.3% | — |
| SciCode | 59.8% | — |
| GSO | — | 3.6% |
| WeirdML | — | 52.6% |
| CadEval | — | 62% |
| AlgoTune | — | 1.72 |

## Agentic & Tool Use

- Gemini 3.7 Flash: 42.1 (#19)
- o4-mini: 32.6 (#61)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| APEX-Agents | 67.8% | — |
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| Remote Labor Index | 5% | — |
| GDP.pdf | 23.8% | — |
| METR Time Horizons | — | 63.9% |

## Reasoning

- Gemini 3.7 Flash: 70.0 (#15)
- o4-mini: 24.6 (#162)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| ARC-AGI-2 | 84.6% | 6.1% |
| ARC-AGI-1 | 95.5% | 58.7% |
| CritPt | 14.3% | 0.6% |
| Chess Puzzles | 47% | 26% |
| LMArena Hard Prompts | 1494 | 1351 |
| Mystery Game Puzzles | 37% | 5% |
| DTBench | 96.8% | 77.6% |
| LMCA | 50.4% | 26.5% |
| Epoch Capabilities Index | 157.27 | 145.64 |
| SimpleBench | — | 38.7% |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 94% | — |
| EnigmaEval | — | 9.2% |
| ForecastBench | — | 61.8 |

## Math

- Gemini 3.7 Flash: 69.6 (#23)
- o4-mini: 40.8 (#89)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 71.6% | 36.1% |
| FrontierMath Tier 4 | 36.6% | 4.9% |
| OTIS Mock AIME 2024-2025 | 97.2% | 81.7% |
| LMArena Math | 1507 | 1389 |
| ProofBench | 58% | — |
| Omni-MATH | — | 72% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 24.8% |
| FrontierMath Tier 4 (v1) | — | 6.3% |

## Knowledge

- Gemini 3.7 Flash: 69.7 (#5)
- o4-mini: 43.6 (#91)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| GPQA Diamond | 94.8% | 79.6% |
| SimpleQA Verified | 69.2% | 19.6% |
| LMArena Expert | 1508 | 1343 |
| Humanity's Last Exam | — | 18.1% |
| MMLU-Pro | — | 82% |
| Confabulations | — | 15.8% |
| Vectara Hallucination Rate | — | 18.6% |
| GPQA (HELM) | — | 73.5% |

## Multimodal

- Gemini 3.7 Flash: 37.3 (#73)
- o4-mini: 40.2 (#49)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| LMArena Vision | 1316 | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |
| Furniture Assembly | 26.7% | — |

## Multilingual

- Gemini 3.7 Flash: 57.6 (#7)
- o4-mini: 47.0 (#154)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| LMArena Non-English | 1484 | 1337 |
| LMArena Chinese | 1548 | 1354 |
| LMArena French | 1505 | 1364 |
| LMArena German | 1498 | 1336 |
| LMArena Japanese | 1512 | 1308 |
| LMArena Korean | 1483 | 1312 |
| LMArena Russian | 1516 | 1334 |
| LMArena Spanish | 1503 | 1347 |

## Instruction Following

- Gemini 3.7 Flash: 77.7 (#15)
- o4-mini: 75.2 (#68)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1483 | 1321 |
| IFEval | — | 92.8% |

## Long Context

- Gemini 3.7 Flash: 45.7 (#30)
- o4-mini: 45.5 (#33)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| LMArena Longer Query | 1492 | 1315 |
| Fiction.LiveBench | — | 77.8% |

## Writing & Preference

- Gemini 3.7 Flash: 71.2 (#20)
- o4-mini: 54.0 (#152)

| Benchmark | Gemini 3.7 Flash | o4-mini |
|---|---|---|
| LMArena Text | 1486 | 1353 |
| LMArena Creative Writing | 1490 | 1294 |
| LMArena Multi-Turn | 1489 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 1723 | — |
| WildBench | — | 85.4% |

## FAQ

### Is Gemini 3.7 Flash better than o4-mini?

Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 41.6 on the Noometry Index.

### Which is cheaper, Gemini 3.7 Flash or o4-mini?

Gemini 3.7 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; o4-mini lists at $1.10 and $4.40.

### Is Gemini 3.7 Flash or o4-mini better for coding?

Gemini 3.7 Flash scores higher on coding benchmarks: 56.2 versus 40.9 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.7 Flash does, with 1.05M tokens against 200K.

### How many benchmarks do Gemini 3.7 Flash and o4-mini share?

32 benchmarks have published results for both models. Gemini 3.7 Flash has 44 scored results on Noometry and o4-mini has 60.