# Gemma 3 27B vs GPT-4.1 mini

> GPT-4.1 mini is the stronger model overall, scoring 33.5 to 30.5 on the Noometry Index. Gemma 3 27B costs 7.0× less per token, which makes it the better buy when GPT-4.1 mini's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemma-3-27b-vs-gpt-4-1-mini
- Last updated: 2026-10-10
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. Gemma 3 27B scores higher in 4 categories and GPT-4.1 mini in 6 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-4.1 mini leads 34.7 to 25.4.
- The biggest single-benchmark swing is Aider Polyglot: 4.9% for Gemma 3 27B and 32.4% for GPT-4.1 mini.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.40 / $1.60 for GPT-4.1 mini.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 131K.
- Gemma 3 27B has downloadable open weights; the other is API-only.

## Snapshot

| | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 30.5 | 33.5 |
| Rank | 277 | 232 |
| Context | 131K | 1.05M |
| Input $/M | $0.08 | $0.40 |
| Output $/M | $0.16 | $1.60 |
| Weights | Open | Proprietary |

## Coding

- Gemma 3 27B: 22.1 (#329)
- GPT-4.1 mini: 30.4 (#286)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| Aider Polyglot | 4.9% | 32.4% |
| SciCode | 21.2% | 40.4% |
| LMArena Coding | 1322 | 1367 |
| SWE-bench Verified (bash only) | — | 23.9% |
| WeirdML | — | 37.6% |
| BigCodeBench Instruct | — | 48.9% |
| LiveBench Coding | 39.9% | — |
| CadEval | — | 16% |

## Agentic & Tool Use

- Gemma 3 27B: 24.9 (#105)
- GPT-4.1 mini: 33.0 (#52)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | 50.5% |

## Reasoning

- Gemma 3 27B: 16.3 (#297)
- GPT-4.1 mini: 10.9 (#335)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| Kagi LLM Benchmark | 40.4% | 48.6% |
| CritPt | 0% | 0% |
| Chess Puzzles | 0% | 7% |
| LMArena Hard Prompts | 1340 | 1349 |
| DTBench | 52.5% | 68.8% |
| LMCA | 12.3% | 21.1% |
| Epoch Capabilities Index | 130.04 | 135.01 |
| ARC-AGI-2 | — | 0% |
| ARC-AGI-1 | — | 3.5% |
| LiveBench Reasoning | 43.8% | — |
| Mystery Game Puzzles | — | 7% |
| LiveBench Data Analysis | 51.5% | — |
| LiveBench | 50% | — |

## Math

- Gemma 3 27B: 24.4 (#261)
- GPT-4.1 mini: 23.1 (#264)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 44.7% |
| LMArena Math | 1312 | 1343 |
| MATH Level 5 | 74% | 87.3% |
| FrontierMath (Tiers 1-3) | — | 6.7% |
| Omni-MATH | — | 49.1% |
| LiveBench Math | 55.4% | — |
| FrontierMath (Feb 2025 set) | — | 4.5% |

## Knowledge

- Gemma 3 27B: 25.4 (#256)
- GPT-4.1 mini: 34.7 (#183)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| GPQA Diamond | 47.7% | 65.8% |
| LMArena Expert | 1303 | 1338 |
| SimpleQA Verified | — | 12.7% |
| MMLU-Pro | — | 78.3% |
| Confabulations | 40.3% | — |
| Vectara Hallucination Rate | 7.4% | — |
| GPQA (HELM) | — | 61.4% |

## Multimodal

- Gemma 3 27B: 32.0 (#94)
- GPT-4.1 mini: 35.2 (#77)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| LMArena Vision | 1165 | 1181 |
| GeoBench | 52% | — |

## Multilingual

- Gemma 3 27B: 47.0 (#150)
- GPT-4.1 mini: 45.9 (#161)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| LMArena Non-English | 1334 | 1318 |
| LMArena Chinese | 1347 | 1329 |
| LMArena French | 1368 | 1357 |
| LMArena German | 1362 | 1349 |
| LMArena Japanese | 1287 | 1291 |
| LMArena Korean | 1308 | 1298 |
| LMArena Russian | 1349 | 1324 |
| LMArena Spanish | 1348 | 1319 |

## Instruction Following

- Gemma 3 27B: 70.8 (#155)
- GPT-4.1 mini: 73.9 (#113)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| LMArena Instruction Following | 1321 | 1332 |
| LiveBench Instruction Following | 74.9% | — |
| IFEval | — | 90.4% |

## Long Context

- Gemma 3 27B: 27.6 (#288)
- GPT-4.1 mini: 31.8 (#270)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| Fiction.LiveBench | 33.3% | 44.4% |
| LMArena Longer Query | 1333 | 1344 |

## Writing & Preference

- Gemma 3 27B: 52.7 (#161)
- GPT-4.1 mini: 48.8 (#193)

| Benchmark | Gemma 3 27B | GPT-4.1 mini |
|---|---|---|
| LMArena Text | 1358 | 1341 |
| LMArena Creative Writing | 1346 | 1300 |
| EQ-Bench Creative Writing | 1266 | 1147 |
| LMArena Multi-Turn | 1344 | 1354 |
| Short-Story Creative Writing | 79.9% | — |
| WildBench | — | 83.8% |
| LiveBench Language | 34.6% | — |

## FAQ

### Is Gemma 3 27B better than GPT-4.1 mini?

GPT-4.1 mini is the stronger model overall, scoring 33.5 to 30.5 on the Noometry Index. Gemma 3 27B costs 7.0× less per token, which makes it the better buy when GPT-4.1 mini's lead doesn't matter for your workload.

### Which is cheaper, Gemma 3 27B or GPT-4.1 mini?

Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; GPT-4.1 mini lists at $0.40 and $1.60.

### Is Gemma 3 27B or GPT-4.1 mini better for coding?

GPT-4.1 mini scores higher on coding benchmarks: 30.4 versus 22.1 in the Noometry coding category.

### Which has the bigger context window?

GPT-4.1 mini does, with 1.05M tokens against 131K.

### How many benchmarks do Gemma 3 27B and GPT-4.1 mini share?

32 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and GPT-4.1 mini has 47.