# Gemma 2 27B vs o4-mini

> o4-mini is the stronger model overall, scoring 41.6 to 29.2 on the Noometry Index. Gemma 2 27B costs 3.0× less per token, which makes it the better buy when o4-mini's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemma-2-27b-vs-o4-mini
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. Gemma 2 27B scores higher in 0 categories and o4-mini in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where o4-mini leads 40.2 to 10.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.4% for Gemma 2 27B and 81.7% for o4-mini.
- Gemma 2 27B is cheaper at $0.65 / $0.65 per million input/output tokens, against $1.10 / $4.40 for o4-mini.
- o4-mini accepts more context: 200K tokens versus 8K.
- Gemma 2 27B has downloadable open weights; the other is API-only.

## Snapshot

| | Gemma 2 27B | o4-mini |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 29.2 | 41.6 |
| Rank | 302 | 119 |
| Context | 8K | 200K |
| Input $/M | $0.65 | $1.10 |
| Output $/M | $0.65 | $4.40 |
| Weights | Open | Proprietary |

## Coding

- Gemma 2 27B: 33.5 (#243)
- o4-mini: 41.6 (#110)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| LMArena Coding | 1211 | 1369 |
| SWE-bench Verified (bash only) | — | 45% |
| Aider Polyglot | — | 72% |
| GSO | — | 3.6% |
| WeirdML | — | 52.6% |
| BigCodeBench Instruct | 42.8% | — |
| LiveBench Coding | 36% | — |
| BigCodeBench Complete | 52.5% | — |
| CadEval | — | 62% |
| ALE-Bench | — | 826.17 |
| AlgoTune | — | 1.72 |

## Agentic & Tool Use

- Gemma 2 27B: —
- o4-mini: 32.3 (#57)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| METR Time Horizons | — | 63.9% |

## Reasoning

- Gemma 2 27B: 14.8 (#313)
- o4-mini: 24.6 (#143)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| LMArena Hard Prompts | 1199 | 1351 |
| DTBench | 48% | 77.6% |
| LMCA | 7.1% | 26.5% |
| Epoch Capabilities Index | 122.06 | 145.63 |
| ARC-AGI-2 | — | 6.1% |
| SimpleBench | — | 38.7% |
| Kagi LLM Benchmark | — | 67.6% |
| ARC-AGI-1 | — | 58.7% |
| CritPt | — | 0.6% |
| Chess Puzzles | — | 26% |
| EnigmaEval | — | 9.2% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 5% |
| LiveBench Data Analysis | 47.9% | — |
| ForecastBench | — | 61.8 |
| LiveBench | 38.2% | — |

## Math

- Gemma 2 27B: 10.0 (#306)
- o4-mini: 40.2 (#79)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 81.7% |
| LMArena Math | 1212 | 1390 |
| MATH Level 5 | 27.9% | 97.8% |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| Omni-MATH | — | 72% |
| LiveBench Math | 26.5% | — |
| FrontierMath (Feb 2025 set) | — | 24.8% |
| FrontierMath Tier 4 (v1) | — | 6.3% |

## Knowledge

- Gemma 2 27B: 18.9 (#275)
- o4-mini: 43.5 (#83)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| GPQA Diamond | 36.5% | 79.6% |
| Confabulations | 27.1% | 15.8% |
| LMArena Expert | 1172 | 1344 |
| Humanity's Last Exam | — | 18.1% |
| SimpleQA Verified | — | 19.6% |
| MMLU-Pro | — | 82% |
| Vectara Hallucination Rate | — | 18.6% |
| GPQA (HELM) | — | 73.5% |
| MMLU | 75.7% | — |

## Multimodal

- Gemma 2 27B: —
- o4-mini: 39.5 (#46)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| LMArena Vision | — | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |

## Multilingual

- Gemma 2 27B: 38.8 (#221)
- o4-mini: 47.2 (#149)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| LMArena Non-English | 1217 | 1336 |
| LMArena Chinese | 1221 | 1355 |
| LMArena French | 1246 | 1362 |
| LMArena German | 1209 | 1336 |
| LMArena Japanese | 1175 | 1308 |
| LMArena Korean | 1174 | 1313 |
| LMArena Russian | 1234 | 1334 |
| LMArena Spanish | 1228 | 1346 |

## Instruction Following

- Gemma 2 27B: 60.7 (#244)
- o4-mini: 75.4 (#63)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1206 | 1321 |
| LiveBench Instruction Following | 58.1% | — |
| IFEval | — | 92.8% |

## Long Context

- Gemma 2 27B: 37.3 (#213)
- o4-mini: 45.5 (#31)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| LMArena Longer Query | 1231 | 1315 |
| Fiction.LiveBench | — | 77.8% |

## Writing & Preference

- Gemma 2 27B: 44.4 (#220)
- o4-mini: 54.2 (#148)

| Benchmark | Gemma 2 27B | o4-mini |
|---|---|---|
| LMArena Text | 1231 | 1353 |
| LMArena Creative Writing | 1241 | 1294 |
| LMArena Multi-Turn | 1224 | 1350 |
| Short-Story Creative Writing | — | 75% |
| WildBench | — | 85.4% |
| LiveBench Language | 32.6% | — |

## FAQ

### Is Gemma 2 27B better than o4-mini?

o4-mini is the stronger model overall, scoring 41.6 to 29.2 on the Noometry Index. Gemma 2 27B costs 3.0× less per token, which makes it the better buy when o4-mini's lead doesn't matter for your workload.

### Which is cheaper, Gemma 2 27B or o4-mini?

Gemma 2 27B is cheaper. It lists at $0.65 per million input tokens and $0.65 per million output tokens; o4-mini lists at $1.10 and $4.40.

### Is Gemma 2 27B or o4-mini better for coding?

o4-mini scores higher on coding benchmarks: 41.6 versus 33.5 in the Noometry coding category.

### Which has the bigger context window?

o4-mini does, with 200K tokens against 8K.

### How many benchmarks do Gemma 2 27B and o4-mini share?

24 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and o4-mini has 60.