# Gemini 3.7 Flash vs GPT-5.4 Pro

> Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 58.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-7-flash-vs-gpt-5-4-pro
- Last updated: 2026-10-10
- Shared benchmarks: 9

## Summary

- They share 9 benchmarks with published results for both. Gemini 3.7 Flash scores higher in 2 categories and GPT-5.4 Pro in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in coding, where Gemini 3.7 Flash leads 55.7 to 42.6.
- The biggest single-benchmark swing is SimpleQA Verified: 69.2% for Gemini 3.7 Flash and 46.3% for GPT-5.4 Pro.
- Gemini 3.7 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $30 / $180 for GPT-5.4 Pro.
- GPT-5.4 Pro accepts more context: 1.05M tokens versus 1.05M.

## Snapshot

| | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 59.8 | 58.8 |
| Rank | 11 | 16 |
| Context | 1.05M | 1.05M |
| Input $/M | $0.75 | $30 |
| Output $/M | $3.75 | $180 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.7 Flash: 55.7 (#19)
- GPT-5.4 Pro: 42.6 (#89)

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| DeepSWE | 65.5% | — |
| FrontierCode | 43.6% | — |
| LMArena WebDev | 1593 | — |
| FrontierSWE | 20.3% | — |
| SciCode | 59.8% | — |
| WeirdML | — | 57.4% |
| LMArena Coding | 1499 | — |
| ALE-Bench | 904.3 | — |

## Agentic & Tool Use

- Gemini 3.7 Flash: 42.2 (#18)
- GPT-5.4 Pro: —

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| APEX-Agents | 67.8% | — |
| Remote Labor Index | 5% | — |
| GDP.pdf | 23.8% | — |

## Reasoning

- Gemini 3.7 Flash: 70.1 (#12)
- GPT-5.4 Pro: 71.0 (#10)

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| ARC-AGI-2 | 84.6% | 83.3% |
| ARC-AGI-1 | 95.5% | 94.5% |
| CritPt | 14.3% | 30% |
| Chess Puzzles | 47% | 58.6% |
| Epoch Capabilities Index | 157.72 | 159.12 |
| SimpleBench | — | 74.1% |
| NYT Connections (extended) | 94% | — |
| EnigmaEval | — | 23.8% |
| LMArena Hard Prompts | 1497 | — |
| Mystery Game Puzzles | 37% | — |
| DTBench | 96.8% | — |
| LMCA | 50.4% | — |

## Math

- Gemini 3.7 Flash: 69.5 (#18)
- GPT-5.4 Pro: 72.0 (#17)

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 71.6% | 82.5% |
| FrontierMath Tier 4 | 36.6% | 58.5% |
| OTIS Mock AIME 2024-2025 | 97.2% | — |
| ProofBench | 58% | — |
| LMArena Math | 1511 | — |
| FrontierMath (Feb 2025 set) | — | 50% |
| FrontierMath Tier 4 (v1) | — | 37.5% |

## Knowledge

- Gemini 3.7 Flash: 69.8 (#4)
- GPT-5.4 Pro: 68.3 (#6)

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| GPQA Diamond | 94.8% | 94.6% |
| SimpleQA Verified | 69.2% | 46.3% |
| Humanity's Last Exam | — | 44.3% |
| Vectara Hallucination Rate | — | 8.3% |
| LMArena Expert | 1508 | — |

## Multimodal

- Gemini 3.7 Flash: 36.9 (#69)
- GPT-5.4 Pro: —

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Vision | 1314 | — |
| Furniture Assembly | 26.7% | — |

## Multilingual

- Gemini 3.7 Flash: 57.8 (#7)
- GPT-5.4 Pro: —

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Non-English | 1485 | — |
| LMArena Chinese | 1553 | — |
| LMArena French | 1512 | — |
| LMArena German | 1485 | — |
| LMArena Japanese | 1506 | — |
| LMArena Korean | 1486 | — |
| LMArena Russian | 1519 | — |
| LMArena Spanish | 1507 | — |

## Instruction Following

- Gemini 3.7 Flash: 78.0 (#12)
- GPT-5.4 Pro: —

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Instruction Following | 1486 | — |

## Long Context

- Gemini 3.7 Flash: 45.8 (#28)
- GPT-5.4 Pro: —

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Longer Query | 1493 | — |

## Writing & Preference

- Gemini 3.7 Flash: 71.4 (#19)
- GPT-5.4 Pro: —

| Benchmark | Gemini 3.7 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Text | 1487 | — |
| LMArena Creative Writing | 1490 | — |
| EQ-Bench Creative Writing | 1723 | — |
| LMArena Multi-Turn | 1495 | — |

## FAQ

### Is Gemini 3.7 Flash better than GPT-5.4 Pro?

Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 58.8 on the Noometry Index.

### Which is cheaper, Gemini 3.7 Flash or GPT-5.4 Pro?

Gemini 3.7 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; GPT-5.4 Pro lists at $30 and $180.

### Is Gemini 3.7 Flash or GPT-5.4 Pro better for coding?

Gemini 3.7 Flash scores higher on coding benchmarks: 55.7 versus 42.6 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.4 Pro does, with 1.05M tokens against 1.05M.

### How many benchmarks do Gemini 3.7 Flash and GPT-5.4 Pro share?

9 benchmarks have published results for both models. Gemini 3.7 Flash has 44 scored results on Noometry and GPT-5.4 Pro has 16.