# Gemma 3 12B vs Phi 3 Small 8k Instruct

> Gemma 3 12B is the stronger model overall, scoring 31.8 to 28.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemma-3-12b-vs-phi-3-small-8k-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 12

## Summary

- They share 12 benchmarks with published results for both. Gemma 3 12B scores higher in 6 categories and Phi 3 Small 8k Instruct in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemma 3 12B leads 45.9 to 28.6.

## Snapshot

| | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| Provider | Google | Microsoft |
| Noometry Index | 31.8 | 28.8 |
| Rank | 255 | 310 |
| Context | 131K | — |
| Input $/M | $0.05 | — |
| Output $/M | $0.15 | — |
| Weights | Open | Open |

## Coding

- Gemma 3 12B: 31.2 (#276)
- Phi 3 Small 8k Instruct: 27.4 (#313)

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Coding | 1281 | 1102 |
| SciCode | 17.4% | — |
| LiveBench Coding | — | 20.3% |

## Agentic & Tool Use

- Gemma 3 12B: 25.3 (#101)
- Phi 3 Small 8k Instruct: —

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 30.4% | — |

## Reasoning

- Gemma 3 12B: 15.2 (#307)
- Phi 3 Small 8k Instruct: 14.2 (#319)

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1309 | 1100 |
| CritPt | 0% | — |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 15.9% |
| DTBench | 48.8% | — |
| LiveBench Data Analysis | — | 30.3% |
| LMCA | 4.5% | — |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 79.1% |
| Epoch Capabilities Index | 123.49 | — |
| HellaSwag | — | 77% |
| LiveBench | — | 24% |
| WinoGrande | — | 81.5% |

## Math

- Gemma 3 12B: 21.0 (#273)
- Phi 3 Small 8k Instruct: 25.4 (#247)

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Math | 1307 | 1151 |
| OTIS Mock AIME 2024-2025 | 16.7% | — |
| LiveBench Math | — | 17.6% |

## Knowledge

- Gemma 3 12B: 26.5 (#250)
- Phi 3 Small 8k Instruct: 28.5 (#237)

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Expert | 1248 | 1067 |
| GPQA Diamond | 39.5% | — |
| Vectara Hallucination Rate | 4.4% | — |
| ARC (AI2) Challenge | — | 90.7% |
| MMLU | — | 75.7% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 58.1% |

## Multimodal

- Gemma 3 12B: —
- Phi 3 Small 8k Instruct: —

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| MindCube | 46.7% | — |

## Multilingual

- Gemma 3 12B: 45.9 (#160)
- Phi 3 Small 8k Instruct: 28.6 (#267)

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Non-English | 1318 | 1058 |
| LMArena German | 1370 | 1080 |
| LMArena Russian | 1335 | 1111 |
| LMArena Chinese | — | 1061 |
| LMArena French | — | 1134 |
| LMArena Japanese | — | 966 |
| LMArena Korean | — | 894 |
| LMArena Spanish | — | 1111 |

## Instruction Following

- Gemma 3 12B: 68.8 (#179)
- Phi 3 Small 8k Instruct: 52.2 (#287)

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Instruction Following | 1299 | 1087 |
| LiveBench Instruction Following | — | 47.2% |

## Long Context

- Gemma 3 12B: 40.0 (#157)
- Phi 3 Small 8k Instruct: 33.0 (#262)

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Longer Query | 1317 | 1088 |

## Writing & Preference

- Gemma 3 12B: 47.7 (#205)
- Phi 3 Small 8k Instruct: 31.3 (#279)

| Benchmark | Gemma 3 12B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Text | 1334 | 1110 |
| LMArena Creative Writing | 1331 | 1083 |
| LMArena Multi-Turn | 1334 | 1068 |
| EQ-Bench Creative Writing | 1126 | — |
| LiveBench Language | — | 12.9% |

## FAQ

### Is Gemma 3 12B better than Phi 3 Small 8k Instruct?

Gemma 3 12B is the stronger model overall, scoring 31.8 to 28.8 on the Noometry Index.

### Is Gemma 3 12B or Phi 3 Small 8k Instruct better for coding?

Gemma 3 12B scores higher on coding benchmarks: 31.2 versus 27.4 in the Noometry coding category.

### How many benchmarks do Gemma 3 12B and Phi 3 Small 8k Instruct share?

12 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and Phi 3 Small 8k Instruct has 32.