# Gemini 2.5 Flash vs GPT-5.4 nano

> GPT-5.4 nano is the stronger model overall, scoring 41.9 to 39.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-2-5-flash-vs-gpt-5-4-nano
- Last updated: 2026-10-12
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 4 categories and GPT-5.4 nano in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where GPT-5.4 nano leads 43.6 to 35.8.
- The biggest single-benchmark swing is ARC-AGI-1: 33.3% for Gemini 2.5 Flash and 51.5% for GPT-5.4 nano.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $0.30 / $2.50 for Gemini 2.5 Flash.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 400K.

## Snapshot

| | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 39.3 | 41.9 |
| Rank | 170 | 125 |
| Context | 1.05M | 400K |
| Input $/M | $0.30 | $0.20 |
| Output $/M | $2.50 | $1.25 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 2.5 Flash: 35.8 (#220)
- GPT-5.4 nano: 43.6 (#84)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| WeirdML | 41.9% | 49.2% |
| LMArena Coding | 1424 | 1405 |
| ALE-Bench | 661.88 | 1,005 |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| SciCode | — | 46.9% |

## Agentic & Tool Use

- Gemini 2.5 Flash: 30.8 (#74)
- GPT-5.4 nano: —

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| Terminal-Bench | 17.1% | — |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
| Vending-Bench 2 | 548.84 | — |

## Reasoning

- Gemini 2.5 Flash: 18.1 (#286)
- GPT-5.4 nano: 23.7 (#173)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| ARC-AGI-2 | 2.5% | 5.7% |
| Kagi LLM Benchmark | 56.8% | 39.7% |
| ARC-AGI-1 | 33.3% | 51.5% |
| CritPt | 1.1% | 9.3% |
| LMArena Hard Prompts | 1422 | 1381 |
| DTBench | 76.5% | 80.3% |
| LMCA | 27.5% | 36.9% |
| Epoch Capabilities Index | 143.03 | 145.81 |
| ForecastBench | 60.6 | 57.3 |
| SimpleBench | 41.2% | — |
| Chess Puzzles | — | 30% |
| EnigmaEval | 2.7% | — |
| Mystery Game Puzzles | — | 9% |

## Math

- Gemini 2.5 Flash: 39.9 (#98)
- GPT-5.4 nano: 40.9 (#88)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 87.8% |
| LMArena Math | 1415 | 1406 |
| FrontierMath (Feb 2025 set) | 4.8% | 25.9% |
| FrontierMath Tier 4 (v1) | 4.2% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 5% |
| Omni-MATH | 38.5% | — |

## Knowledge

- Gemini 2.5 Flash: 36.4 (#168)
- GPT-5.4 nano: 41.9 (#103)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| Vectara Hallucination Rate | 7.8% | 3.1% |
| LMArena Expert | 1426 | 1396 |
| GPQA Diamond | — | 78.5% |
| Humanity's Last Exam | 12.1% | — |
| SimpleQA Verified | — | 11.7% |
| MMLU-Pro | 63.9% | — |
| Confabulations | 16.8% | — |
| GPQA (HELM) | 39% | — |

## Multimodal

- Gemini 2.5 Flash: 41.8 (#32)
- GPT-5.4 nano: 36.7 (#78)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | 1253 | 1196 |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |

## Multilingual

- Gemini 2.5 Flash: 52.3 (#88)
- GPT-5.4 nano: 48.6 (#140)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1409 | 1359 |
| LMArena Chinese | 1450 | 1392 |
| LMArena French | 1433 | 1396 |
| LMArena German | 1418 | 1367 |
| LMArena Japanese | 1405 | 1343 |
| LMArena Korean | 1385 | 1320 |
| LMArena Russian | 1415 | 1363 |
| LMArena Spanish | 1421 | 1371 |

## Instruction Following

- Gemini 2.5 Flash: 75.7 (#54)
- GPT-5.4 nano: 71.9 (#144)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1405 | 1362 |
| IFEval | 89.8% | — |

## Long Context

- Gemini 2.5 Flash: 47.5 (#17)
- GPT-5.4 nano: 41.6 (#137)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1419 | 1366 |
| Fiction.LiveBench | 77.8% | — |

## Writing & Preference

- Gemini 2.5 Flash: 53.8 (#157)
- GPT-5.4 nano: 55.7 (#142)

| Benchmark | Gemini 2.5 Flash | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1417 | 1372 |
| LMArena Creative Writing | 1400 | 1314 |
| LMArena Multi-Turn | 1408 | 1382 |
| Short-Story Creative Writing | 76.5% | — |
| EQ-Bench Creative Writing | 1137 | — |
| WildBench | 81.7% | — |

## FAQ

### Is Gemini 2.5 Flash better than GPT-5.4 nano?

GPT-5.4 nano is the stronger model overall, scoring 41.9 to 39.3 on the Noometry Index.

### Which is cheaper, Gemini 2.5 Flash or GPT-5.4 nano?

GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; Gemini 2.5 Flash lists at $0.30 and $2.50.

### Is Gemini 2.5 Flash or GPT-5.4 nano better for coding?

GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 35.8 in the Noometry coding category.

### Which has the bigger context window?

Gemini 2.5 Flash does, with 1.05M tokens against 400K.

### How many benchmarks do Gemini 2.5 Flash and GPT-5.4 nano share?

32 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and GPT-5.4 nano has 40.