# Gemma 3 12B vs Llama 2-13B

> Gemma 3 12B is the stronger model overall, scoring 31.8 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemma-3-12b-vs-llama-2-13b
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. Gemma 3 12B scores higher in 6 categories and Llama 2-13B in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemma 3 12B leads 45.9 to 26.7.
- The biggest single-benchmark swing is DTBench: 48.8% for Gemma 3 12B and 42.2% for Llama 2-13B.

## Snapshot

| | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| Provider | Google | Meta |
| Noometry Index | 31.8 | 29.1 |
| Rank | 255 | 304 |
| Context | 131K | — |
| Input $/M | $0.05 | — |
| Output $/M | $0.15 | — |
| Weights | Open | Open |

## Coding

- Gemma 3 12B: 31.2 (#276)
- Llama 2-13B: 30.4 (#287)

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1281 | 1062 |
| SciCode | 17.4% | — |

## Agentic & Tool Use

- Gemma 3 12B: 25.3 (#101)
- Llama 2-13B: —

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 30.4% | — |

## Reasoning

- Gemma 3 12B: 15.2 (#307)
- Llama 2-13B: 12.3 (#333)

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1309 | 1051 |
| DTBench | 48.8% | 42.2% |
| Epoch Capabilities Index | 123.49 | 105.8 |
| CritPt | 0% | — |
| LMCA | 4.5% | — |
| BIG-Bench Hard | — | 58.2% |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| PIQA | — | 80.8% |
| WinoGrande | — | 72.8% |

## Math

- Gemma 3 12B: 21.0 (#273)
- Llama 2-13B: 28.7 (#226)

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| LMArena Math | 1307 | 1065 |
| OTIS Mock AIME 2024-2025 | 16.7% | — |
| GSM8K | — | 36.9% |

## Knowledge

- Gemma 3 12B: 26.5 (#250)
- Llama 2-13B: 27.5 (#244)

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1248 | 1030 |
| GPQA Diamond | 39.5% | — |
| Vectara Hallucination Rate | 4.4% | — |
| ARC (AI2) Challenge | — | 60.3% |
| BoolQ | — | 82.4% |
| MMLU | — | 55.6% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |

## Multimodal

- Gemma 3 12B: —
- Llama 2-13B: —

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| MindCube | 46.7% | — |
| ScienceQA | — | 55.8% |

## Multilingual

- Gemma 3 12B: 45.9 (#160)
- Llama 2-13B: 26.7 (#274)

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1318 | 1024 |
| LMArena German | 1370 | 1009 |
| LMArena Russian | 1335 | 1055 |
| LMArena Chinese | — | 1001 |
| LMArena French | — | 1043 |
| LMArena Japanese | — | 894 |
| LMArena Korean | — | 953 |
| LMArena Spanish | — | 1086 |

## Instruction Following

- Gemma 3 12B: 68.8 (#179)
- Llama 2-13B: 53.5 (#282)

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1299 | 1046 |

## Long Context

- Gemma 3 12B: 40.0 (#157)
- Llama 2-13B: 32.3 (#264)

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1317 | 1064 |

## Writing & Preference

- Gemma 3 12B: 47.7 (#205)
- Llama 2-13B: 30.0 (#284)

| Benchmark | Gemma 3 12B | Llama 2-13B |
|---|---|---|
| LMArena Text | 1334 | 1085 |
| LMArena Creative Writing | 1331 | 1047 |
| LMArena Multi-Turn | 1334 | 1050 |
| EQ-Bench Creative Writing | 1126 | — |

## FAQ

### Is Gemma 3 12B better than Llama 2-13B?

Gemma 3 12B is the stronger model overall, scoring 31.8 to 29.1 on the Noometry Index.

### Is Gemma 3 12B or Llama 2-13B better for coding?

They score almost the same on coding (31.2 vs 30.4); test both on your own repository before choosing.

### How many benchmarks do Gemma 3 12B and Llama 2-13B share?

15 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and Llama 2-13B has 32.