# Gemma 2 9B vs GPT-4

> GPT-4 is the stronger model overall, scoring 28.9 to 25.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemma-2-9b-vs-gpt-4
- Last updated: 2026-10-10
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. Gemma 2 9B scores higher in 0 categories and GPT-4 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-4 leads 18.4 to 9.7.
- The biggest single-benchmark swing is BigCodeBench Complete: 40.6% for Gemma 2 9B and 57.2% for GPT-4.
- Gemma 2 9B has downloadable open weights; the other is API-only.

## Snapshot

| | Gemma 2 9B | GPT-4 |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 25.7 | 28.9 |
| Rank | 334 | 308 |
| Context | — | 8K |
| Input $/M | — | $30 |
| Output $/M | — | $60 |
| Weights | Open | Proprietary |

## Coding

- Gemma 2 9B: 29.0 (#299)
- GPT-4: 31.2 (#277)

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| BigCodeBench Instruct | 34.7% | 46% |
| LMArena Coding | 1174 | 1254 |
| BigCodeBench Complete | 40.6% | 57.2% |
| WeirdML | — | 12.4% |
| LiveBench Coding | 22.5% | — |
| HumanEval+ | — | 79.3% |

## Agentic & Tool Use

- Gemma 2 9B: —
- GPT-4: —

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| METR Time Horizons | — | 36.1% |

## Reasoning

- Gemma 2 9B: 15.3 (#306)
- GPT-4: 17.3 (#285)

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| LMArena Hard Prompts | 1172 | 1241 |
| Epoch Capabilities Index | 119.78 | 125.88 |
| Chess Puzzles | — | 4% |
| LiveBench Reasoning | 15.2% | — |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 62.7% |
| LiveBench Data Analysis | 36.4% | — |
| LMCA | — | 17.1% |
| BIG-Bench Hard | — | 75.1% |
| ForecastBench | — | 57.8 |
| HellaSwag | — | 95.3% |
| LiveBench | 28.7% | — |
| PIQA | 83.7% | — |
| WinoGrande | — | 87.5% |

## Math

- Gemma 2 9B: 9.3 (#312)
- GPT-4: 10.1 (#304)

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.6% | 1.1% |
| LMArena Math | 1183 | 1269 |
| MATH Level 5 | 21% | 23% |
| GSM8K | 84.9% | 92% |
| LiveBench Math | 19.8% | — |

## Knowledge

- Gemma 2 9B: 9.7 (#300)
- GPT-4: 18.4 (#277)

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| GPQA Diamond | 27.5% | 35.7% |
| LMArena Expert | 1147 | 1211 |
| MMLU | 72.1% | 86.4% |
| BoolQ | 85.7% | — |
| TriviaQA | — | 84.8% |

## Multilingual

- Gemma 2 9B: 36.8 (#233)
- GPT-4: 40.8 (#210)

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| LMArena Non-English | 1188 | 1246 |
| LMArena Chinese | 1185 | 1242 |
| LMArena French | 1189 | 1282 |
| LMArena German | 1186 | 1251 |
| LMArena Japanese | 1144 | 1209 |
| LMArena Korean | 1137 | 1184 |
| LMArena Russian | 1200 | 1251 |
| LMArena Spanish | 1200 | 1261 |

## Instruction Following

- Gemma 2 9B: 57.8 (#264)
- GPT-4: 65.5 (#217)

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1178 | 1241 |
| LiveBench Instruction Following | 52.6% | — |

## Long Context

- Gemma 2 9B: 36.3 (#228)
- GPT-4: 37.7 (#207)

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1197 | 1244 |

## Writing & Preference

- Gemma 2 9B: 32.3 (#276)
- GPT-4: 35.0 (#263)

| Benchmark | Gemma 2 9B | GPT-4 |
|---|---|---|
| LMArena Text | 1207 | 1263 |
| LMArena Creative Writing | 1206 | 1244 |
| EQ-Bench Creative Writing | 841 | 752 |
| LMArena Multi-Turn | 1193 | 1257 |
| LiveBench Language | 25.5% | — |

## FAQ

### Is Gemma 2 9B better than GPT-4?

GPT-4 is the stronger model overall, scoring 28.9 to 25.7 on the Noometry Index.

### Is Gemma 2 9B or GPT-4 better for coding?

GPT-4 scores higher on coding benchmarks: 31.2 versus 29.0 in the Noometry coding category.

### How many benchmarks do Gemma 2 9B and GPT-4 share?

26 benchmarks have published results for both models. Gemma 2 9B has 35 scored results on Noometry and GPT-4 has 38.