# Gemma 2B vs Phi-4

> Phi-4 is the stronger model overall, scoring 30.9 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemma-2b-vs-phi-4
- Last updated: 2026-10-10
- Shared benchmarks: 13

## Summary

- They share 13 benchmarks with published results for both. Gemma 2B scores higher in 2 categories and Phi-4 in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Phi-4 leads 40.7 to 24.1.

## Snapshot

| | Gemma 2B | Phi-4 |
|---|---|---|
| Provider | Google | Microsoft |
| Noometry Index | 29.1 | 30.9 |
| Rank | 305 | 269 |
| Context | — | 128K |
| Input $/M | — | $0.07 |
| Output $/M | — | $0.14 |
| Weights | Open | Open |

## Coding

- Gemma 2B: 28.9 (#300)
- Phi-4: 33.8 (#237)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| LMArena Coding | 1010 | 1231 |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
| HumanEval+ | 20.7% | — |
| MBPP+ | 34.1% | — |

## Agentic & Tool Use

- Gemma 2B: —
- Phi-4: 22.6 (#120)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |

## Reasoning

- Gemma 2B: 17.9 (#275)
- Phi-4: 17.0 (#287)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| LMArena Hard Prompts | 989 | 1220 |
| Epoch Capabilities Index | 93.57 | 130.43 |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| BIG-Bench Hard | 35.2% | — |
| HellaSwag | 71.4% | — |
| LiveBench | — | 41.6% |
| PIQA | 77.3% | — |
| WinoGrande | 65.4% | — |

## Math

- Gemma 2B: 27.6 (#237)
- Phi-4: 19.6 (#279)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| LMArena Math | 1009 | 1246 |
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
| GSM8K | 17.7% | — |

## Knowledge

- Gemma 2B: —
- Phi-4: 32.5 (#204)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| MMLU | 42.3% | 84.8% |
| GPQA Diamond | — | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| LMArena Expert | — | 1203 |
| ARC (AI2) Challenge | 42.1% | — |
| BoolQ | 69.4% | — |
| TriviaQA | 53.2% | — |

## Multilingual

- Gemma 2B: 23.1 (#289)
- Phi-4: 37.4 (#232)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| LMArena Non-English | 958 | 1197 |
| LMArena Chinese | 986 | 1211 |
| LMArena Russian | 937 | 1209 |
| LMArena French | — | 1223 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1158 |
| LMArena Korean | — | 1151 |
| LMArena Spanish | — | 1234 |

## Instruction Following

- Gemma 2B: 48.8 (#297)
- Phi-4: 60.6 (#246)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 971 | 1201 |
| LiveBench Instruction Following | — | 58.4% |

## Long Context

- Gemma 2B: 29.9 (#286)
- Phi-4: 36.9 (#221)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| LMArena Longer Query | 981 | 1217 |

## Writing & Preference

- Gemma 2B: 24.1 (#303)
- Phi-4: 40.7 (#239)

| Benchmark | Gemma 2B | Phi-4 |
|---|---|---|
| LMArena Text | 1002 | 1217 |
| LMArena Creative Writing | 986 | 1182 |
| LMArena Multi-Turn | 945 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| LiveBench Language | — | 25.6% |

## FAQ

### Is Gemma 2B better than Phi-4?

Phi-4 is the stronger model overall, scoring 30.9 to 29.1 on the Noometry Index.

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

Phi-4 scores higher on coding benchmarks: 33.8 versus 28.9 in the Noometry coding category.

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

13 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and Phi-4 has 37.