# Gemma 3 27B vs GPT-5.1

> GPT-5.1 is the stronger model overall, scoring 48.8 to 30.5 on the Noometry Index. Gemma 3 27B costs 34× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemma-3-27b-vs-gpt-5-1
- Last updated: 2026-10-10
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. Gemma 3 27B scores higher in 0 categories and GPT-5.1 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.1 leads 50.3 to 24.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 22.5% for Gemma 3 27B and 88.6% for GPT-5.1.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
- GPT-5.1 accepts more context: 400K tokens versus 131K.
- Gemma 3 27B has downloadable open weights; the other is API-only.

## Snapshot

| | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 30.5 | 48.8 |
| Rank | 277 | 47 |
| Context | 131K | 400K |
| Input $/M | $0.08 | $1.25 |
| Output $/M | $0.16 | $10 |
| Weights | Open | Proprietary |

## Coding

- Gemma 3 27B: 22.1 (#329)
- GPT-5.1: 46.9 (#59)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| SciCode | 21.2% | 43.3% |
| LiveBench Coding | 39.9% | 72.5% |
| LMArena Coding | 1322 | 1454 |
| SWE-bench Verified | — | 68% |
| SWE-bench Verified (bash only) | — | 66% |
| Aider Polyglot | 4.9% | — |
| LMArena WebDev | — | 1395 |
| GSO | — | 13.7% |
| WeirdML | — | 60.8% |
| ALE-Bench | — | 1,192 |

## Agentic & Tool Use

- Gemma 3 27B: 24.9 (#105)
- GPT-5.1: 32.5 (#55)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| Terminal-Bench | — | 47.6% |
| Berkeley Function Calling Leaderboard | 29.5% | — |
| DeepResearch Bench | — | 42.8% |
| LMArena Search | — | 1199 |
| Vending-Bench 2 | — | 1,473 |

## Reasoning

- Gemma 3 27B: 16.3 (#297)
- GPT-5.1: 39.9 (#50)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| CritPt | 0% | 4.9% |
| Chess Puzzles | 0% | 32% |
| LiveBench Reasoning | 43.8% | 95.8% |
| LMArena Hard Prompts | 1340 | 1457 |
| DTBench | 52.5% | 90.1% |
| LiveBench Data Analysis | 51.5% | 72.1% |
| LMCA | 12.3% | 43.9% |
| Epoch Capabilities Index | 130.04 | 149.66 |
| LiveBench | 50% | 78.8% |
| ARC-AGI-2 | — | 17.6% |
| SimpleBench | — | 53.2% |
| Kagi LLM Benchmark | 40.4% | — |
| ARC-AGI-1 | — | 72.8% |
| EnigmaEval | — | 11.2% |
| Mystery Game Puzzles | — | 19% |
| ForecastBench | — | 58.1 |

## Math

- Gemma 3 27B: 24.4 (#261)
- GPT-5.1: 50.3 (#45)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 88.6% |
| LiveBench Math | 55.4% | 94.5% |
| LMArena Math | 1312 | 1446 |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | 74% | — |
| FrontierMath (Feb 2025 set) | — | 31% |
| FrontierMath Tier 4 (v1) | — | 12.5% |

## Knowledge

- Gemma 3 27B: 25.4 (#256)
- GPT-5.1: 50.5 (#65)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 47.7% | 87.6% |
| Vectara Hallucination Rate | 7.4% | 10.9% |
| LMArena Expert | 1303 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| SimpleQA Verified | — | 48% |
| MMLU-Pro | — | 57.9% |
| Confabulations | 40.3% | — |
| GPQA (HELM) | — | 44.2% |

## Multimodal

- Gemma 3 27B: 32.0 (#94)
- GPT-5.1: 44.0 (#17)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1165 | 1250 |
| GeoBench | 52% | — |
| VPCT | — | 58.7% |
| LMArena Document | — | 1403 |

## Multilingual

- Gemma 3 27B: 47.0 (#150)
- GPT-5.1: 54.0 (#51)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1334 | 1431 |
| LMArena Chinese | 1347 | 1495 |
| LMArena French | 1368 | 1451 |
| LMArena German | 1362 | 1439 |
| LMArena Japanese | 1287 | 1453 |
| LMArena Korean | 1308 | 1401 |
| LMArena Russian | 1349 | 1435 |
| LMArena Spanish | 1348 | 1430 |

## Instruction Following

- Gemma 3 27B: 70.8 (#155)
- GPT-5.1: 84.0 (#1)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| LiveBench Instruction Following | 74.9% | 93.3% |
| LMArena Instruction Following | 1321 | 1442 |
| IFEval | — | 93.5% |

## Long Context

- Gemma 3 27B: 27.6 (#288)
- GPT-5.1: 47.6 (#14)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1333 | 1448 |
| Fiction.LiveBench | 33.3% | — |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |

## Writing & Preference

- Gemma 3 27B: 52.7 (#161)
- GPT-5.1: 64.7 (#53)

| Benchmark | Gemma 3 27B | GPT-5.1 |
|---|---|---|
| LMArena Text | 1358 | 1443 |
| LMArena Creative Writing | 1346 | 1427 |
| LMArena Multi-Turn | 1344 | 1449 |
| LiveBench Language | 34.6% | 80.2% |
| Short-Story Creative Writing | 79.9% | — |
| EQ-Bench Creative Writing | 1266 | — |
| WildBench | — | 86.3% |

## FAQ

### Is Gemma 3 27B better than GPT-5.1?

GPT-5.1 is the stronger model overall, scoring 48.8 to 30.5 on the Noometry Index. Gemma 3 27B costs 34× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.

### Which is cheaper, Gemma 3 27B or GPT-5.1?

Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; GPT-5.1 lists at $1.25 and $10.

### Is Gemma 3 27B or GPT-5.1 better for coding?

GPT-5.1 scores higher on coding benchmarks: 46.9 versus 22.1 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.1 does, with 400K tokens against 131K.

### How many benchmarks do Gemma 3 27B and GPT-5.1 share?

34 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and GPT-5.1 has 63.