# Gemini 2.5 Flash vs GPT-4.1

> Gemini 2.5 Flash is the stronger model overall, scoring 39.0 to 35.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-2-5-flash-vs-gpt-4-1
- Last updated: 2026-10-10
- Shared benchmarks: 45

## Summary

- They share 45 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 7 categories and GPT-4.1 in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 2.5 Flash leads 38.1 to 21.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.1% for Gemini 2.5 Flash and 38.3% for GPT-4.1.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $2 / $8 for GPT-4.1.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 1.05M.

## Snapshot

| | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 39.0 | 35.7 |
| Rank | 162 | 215 |
| Context | 1.05M | 1.05M |
| Input $/M | $0.30 | $2 |
| Output $/M | $2.50 | $8 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 2.5 Flash: 35.5 (#210)
- GPT-4.1: 34.2 (#228)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| SWE-bench Verified (bash only) | 28.7% | 39.6% |
| Aider Polyglot | 55.1% | 52.4% |
| WeirdML | 41.9% | 39% |
| LMArena Coding | 1424 | 1391 |
| ALE-Bench | 661.88 | 558.1 |
| SWE-bench Verified | — | 48.5% |
| CadEval | — | 42% |

## Agentic & Tool Use

- Gemini 2.5 Flash: 30.6 (#67)
- GPT-4.1: 34.4 (#40)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 56.2% | 54% |
| Terminal-Bench | 17.1% | — |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
| Vending-Bench 2 | 548.84 | — |

## Reasoning

- Gemini 2.5 Flash: 18.2 (#271)
- GPT-4.1: 11.7 (#334)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| ARC-AGI-2 | 2.5% | 0.4% |
| SimpleBench | 41.2% | 27% |
| Kagi LLM Benchmark | 56.8% | 52.3% |
| ARC-AGI-1 | 33.3% | 5.5% |
| EnigmaEval | 2.7% | 2.2% |
| LMArena Hard Prompts | 1422 | 1384 |
| DTBench | 76.5% | 68.3% |
| LMCA | 27.5% | 25.6% |
| Epoch Capabilities Index | 143.04 | 136.79 |
| ForecastBench | 60.6 | 61.5 |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 6% |

## Math

- Gemini 2.5 Flash: 38.1 (#87)
- GPT-4.1: 21.4 (#269)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 38.3% |
| Omni-MATH | 38.5% | 47.1% |
| LMArena Math | 1415 | 1369 |
| FrontierMath (Feb 2025 set) | 4.8% | 5.5% |
| FrontierMath Tier 4 (v1) | 4.2% | 0% |
| FrontierMath (Tiers 1-3) | — | 6% |
| MATH Level 5 | — | 83% |

## Knowledge

- Gemini 2.5 Flash: 36.1 (#162)
- GPT-4.1: 37.0 (#152)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| Humanity's Last Exam | 12.1% | 5.4% |
| MMLU-Pro | 63.9% | 81.1% |
| Vectara Hallucination Rate | 7.8% | 5.6% |
| GPQA (HELM) | 39% | 65.9% |
| LMArena Expert | 1426 | 1364 |
| GPQA Diamond | — | 66.9% |
| SimpleQA Verified | — | 31.1% |
| Confabulations | 16.8% | — |

## Multimodal

- Gemini 2.5 Flash: 41.2 (#31)
- GPT-4.1: 37.5 (#66)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| LMArena Vision | 1254 | 1210 |
| GeoBench | 76% | 72% |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |

## Multilingual

- Gemini 2.5 Flash: 52.5 (#83)
- GPT-4.1: 49.6 (#128)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1410 | 1370 |
| LMArena Chinese | 1450 | 1382 |
| LMArena French | 1433 | 1383 |
| LMArena German | 1419 | 1380 |
| LMArena Japanese | 1403 | 1320 |
| LMArena Korean | 1385 | 1339 |
| LMArena Russian | 1415 | 1377 |
| LMArena Spanish | 1421 | 1375 |

## Instruction Following

- Gemini 2.5 Flash: 75.9 (#50)
- GPT-4.1: 71.5 (#148)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| IFEval | 89.8% | 83.8% |
| LMArena Instruction Following | 1405 | 1367 |

## Long Context

- Gemini 2.5 Flash: 47.5 (#16)
- GPT-4.1: 40.0 (#158)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| Fiction.LiveBench | 77.8% | 63.9% |
| LMArena Longer Query | 1419 | 1385 |

## Writing & Preference

- Gemini 2.5 Flash: 54.0 (#153)
- GPT-4.1: 57.8 (#120)

| Benchmark | Gemini 2.5 Flash | GPT-4.1 |
|---|---|---|
| LMArena Text | 1417 | 1383 |
| LMArena Creative Writing | 1400 | 1363 |
| EQ-Bench Creative Writing | 1137 | 1420 |
| WildBench | 81.7% | 85.4% |
| LMArena Multi-Turn | 1408 | 1398 |
| Short-Story Creative Writing | 76.5% | — |

## FAQ

### Is Gemini 2.5 Flash better than GPT-4.1?

Gemini 2.5 Flash is the stronger model overall, scoring 39.0 to 35.7 on the Noometry Index.

### Which is cheaper, Gemini 2.5 Flash or GPT-4.1?

Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; GPT-4.1 lists at $2 and $8.

### Is Gemini 2.5 Flash or GPT-4.1 better for coding?

Gemini 2.5 Flash scores higher on coding benchmarks: 35.5 versus 34.2 in the Noometry coding category.

### Which has the bigger context window?

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

### How many benchmarks do Gemini 2.5 Flash and GPT-4.1 share?

45 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and GPT-4.1 has 52.