# Gemma 3 4B vs Llama 3.1-405B

> Llama 3.1-405B is the stronger model overall, scoring 30.7 to 28.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemma-3-4b-vs-llama-3-1-405b
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Gemma 3 4B scores higher in 4 categories and Llama 3.1-405B in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 3.1-405B leads 30.4 to 11.8.
- The biggest single-benchmark swing is GPQA Diamond: 23.2% for Gemma 3 4B and 50.9% for Llama 3.1-405B.

## Snapshot

| | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| Provider | Google | Meta |
| Noometry Index | 28.1 | 30.7 |
| Rank | 326 | 288 |
| Context | 131K | — |
| Input $/M | $0.04 | — |
| Output $/M | $0.08 | — |
| Weights | Open | Open |

## Coding

- Gemma 3 4B: 35.9 (#215)
- Llama 3.1-405B: 33.1 (#262)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| LMArena Coding | 1230 | 1291 |
| WeirdML | — | 21.4% |

## Agentic & Tool Use

- Gemma 3 4B: 20.9 (#142)
- Llama 3.1-405B: 21.0 (#140)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |
| TheAgentCompany | — | 7.4% |
| Cybench | — | 7.5% |

## Reasoning

- Gemma 3 4B: 13.2 (#335)
- Llama 3.1-405B: 16.8 (#300)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| Kagi LLM Benchmark | 25.2% | 45% |
| LMArena Hard Prompts | 1253 | 1269 |
| DTBench | 50.9% | 61.4% |
| Epoch Capabilities Index | 116.02 | 128.75 |
| SimpleBench | — | 23% |
| Chess Puzzles | 0% | — |
| LMCA | 2.8% | — |
| BIG-Bench Hard | — | 82.9% |
| ForecastBench | — | 59.9 |
| HellaSwag | — | 89.2% |
| PIQA | — | 85.9% |
| WinoGrande | — | 89.2% |

## Math

- Gemma 3 4B: 16.8 (#292)
- Llama 3.1-405B: 18.4 (#290)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.5% | 9.7% |
| LMArena Math | 1239 | 1281 |
| Omni-MATH | — | 24.9% |
| MATH Level 5 | — | 49.8% |

## Knowledge

- Gemma 3 4B: 11.8 (#299)
- Llama 3.1-405B: 30.4 (#227)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| GPQA Diamond | 23.2% | 50.9% |
| LMArena Expert | 1223 | 1243 |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 17.6% |
| Vectara Hallucination Rate | 6.4% | — |
| GPQA (HELM) | — | 52.2% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 84.5% |
| TriviaQA | — | 82.7% |

## Multilingual

- Gemma 3 4B: 42.5 (#194)
- Llama 3.1-405B: 40.7 (#214)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| LMArena Non-English | 1273 | 1248 |
| LMArena German | 1281 | 1252 |
| LMArena Russian | 1294 | 1265 |
| LMArena Chinese | — | 1242 |
| LMArena French | — | 1279 |
| LMArena Japanese | — | 1208 |
| LMArena Korean | — | 1184 |
| LMArena Spanish | — | 1260 |

## Instruction Following

- Gemma 3 4B: 65.2 (#225)
- Llama 3.1-405B: 65.9 (#214)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| LMArena Instruction Following | 1239 | 1259 |
| IFEval | — | 81.1% |

## Long Context

- Gemma 3 4B: 38.7 (#194)
- Llama 3.1-405B: 38.4 (#197)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| LMArena Longer Query | 1273 | 1266 |

## Writing & Preference

- Gemma 3 4B: 42.0 (#239)
- Llama 3.1-405B: 38.9 (#251)

| Benchmark | Gemma 3 4B | Llama 3.1-405B |
|---|---|---|
| LMArena Text | 1291 | 1284 |
| LMArena Creative Writing | 1271 | 1262 |
| EQ-Bench Creative Writing | 1068 | 870 |
| LMArena Multi-Turn | 1255 | 1297 |
| WildBench | — | 78.3% |

## FAQ

### Is Gemma 3 4B better than Llama 3.1-405B?

Llama 3.1-405B is the stronger model overall, scoring 30.7 to 28.1 on the Noometry Index.

### Is Gemma 3 4B or Llama 3.1-405B better for coding?

Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 33.1 in the Noometry coding category.

### How many benchmarks do Gemma 3 4B and Llama 3.1-405B share?

18 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Llama 3.1-405B has 42.