# Gemma 3 4B vs Phi 3 Small 8k Instruct

> Phi 3 Small 8k Instruct is the stronger model overall, scoring 29.3 to 28.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemma-3-4b-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 4B scores higher in 5 categories and Phi 3 Small 8k Instruct in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi 3 Small 8k Instruct leads 29.1 to 11.8.

## Snapshot

| | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| Provider | Google | Microsoft |
| Noometry Index | 28.1 | 29.3 |
| Rank | 326 | 314 |
| Context | 131K | — |
| Input $/M | $0.04 | — |
| Output $/M | $0.08 | — |
| Weights | Open | Open |

## Coding

- Gemma 3 4B: 35.9 (#215)
- Phi 3 Small 8k Instruct: 27.9 (#318)

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Coding | 1230 | 1101 |
| LiveBench Coding | — | 20.3% |

## Agentic & Tool Use

- Gemma 3 4B: 20.9 (#142)
- Phi 3 Small 8k Instruct: —

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |

## Reasoning

- Gemma 3 4B: 13.2 (#335)
- Phi 3 Small 8k Instruct: 14.8 (#323)

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1100 |
| Kagi LLM Benchmark | 25.2% | — |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 15.9% |
| DTBench | 50.9% | — |
| LiveBench Data Analysis | — | 30.3% |
| LMCA | 2.8% | — |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 79.1% |
| Epoch Capabilities Index | 116.02 | — |
| HellaSwag | — | 77% |
| LiveBench | — | 24% |
| WinoGrande | — | 81.5% |

## Math

- Gemma 3 4B: 16.8 (#292)
- Phi 3 Small 8k Instruct: 27.6 (#248)

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Math | 1239 | 1151 |
| OTIS Mock AIME 2024-2025 | 7.5% | — |
| LiveBench Math | — | 17.6% |

## Knowledge

- Gemma 3 4B: 11.8 (#299)
- Phi 3 Small 8k Instruct: 29.1 (#240)

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Expert | 1223 | 1067 |
| GPQA Diamond | 23.2% | — |
| Vectara Hallucination Rate | 6.4% | — |
| ARC (AI2) Challenge | — | 90.7% |
| MMLU | — | 75.7% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 58.1% |

## Multilingual

- Gemma 3 4B: 42.5 (#194)
- Phi 3 Small 8k Instruct: 28.5 (#272)

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Non-English | 1273 | 1058 |
| LMArena German | 1281 | 1080 |
| LMArena Russian | 1294 | 1111 |
| LMArena Chinese | — | 1061 |
| LMArena French | — | 1135 |
| LMArena Japanese | — | 966 |
| LMArena Korean | — | 894 |
| LMArena Spanish | — | 1111 |

## Instruction Following

- Gemma 3 4B: 65.2 (#225)
- Phi 3 Small 8k Instruct: 51.9 (#292)

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Instruction Following | 1239 | 1087 |
| LiveBench Instruction Following | — | 47.2% |

## Long Context

- Gemma 3 4B: 38.7 (#194)
- Phi 3 Small 8k Instruct: 33.0 (#267)

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Longer Query | 1273 | 1088 |

## Writing & Preference

- Gemma 3 4B: 42.0 (#239)
- Phi 3 Small 8k Instruct: 31.1 (#284)

| Benchmark | Gemma 3 4B | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Text | 1291 | 1110 |
| LMArena Creative Writing | 1271 | 1083 |
| LMArena Multi-Turn | 1255 | 1068 |
| EQ-Bench Creative Writing | 1068 | — |
| LiveBench Language | — | 12.9% |

## FAQ

### Is Gemma 3 4B better than Phi 3 Small 8k Instruct?

Phi 3 Small 8k Instruct is the stronger model overall, scoring 29.3 to 28.1 on the Noometry Index.

### Is Gemma 3 4B or Phi 3 Small 8k Instruct better for coding?

Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 27.9 in the Noometry coding category.

### How many benchmarks do Gemma 3 4B and Phi 3 Small 8k Instruct share?

12 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Phi 3 Small 8k Instruct has 32.
