# Llama 3.1-405B vs phi-3-medium 14B

> Llama 3.1-405B is the stronger model overall, scoring 30.4 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/llama-3-1-405b-vs-phi-3-medium-14b
- Last updated: 2026-10-10
- Shared benchmarks: 9

## Summary

- They share 9 benchmarks with published results for both. Llama 3.1-405B scores higher in 1 category and phi-3-medium 14B in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 3.1-405B leads 30.4 to 9.1.
- The biggest single-benchmark swing is MATH Level 5: 49.8% for Llama 3.1-405B and 17.6% for phi-3-medium 14B.

## Snapshot

| | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| Provider | Meta | Microsoft |
| Noometry Index | 30.4 | 29.1 |
| Rank | 279 | 306 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Llama 3.1-405B: 32.7 (#257)
- phi-3-medium 14B: 36.3 (#198)

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| WeirdML | 21.4% | — |
| BigCodeBench Instruct | — | 37.6% |
| LMArena Coding | 1291 | — |
| BigCodeBench Complete | — | 48.7% |

## Agentic & Tool Use

- Llama 3.1-405B: 20.9 (#131)
- phi-3-medium 14B: —

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| TheAgentCompany | 7.4% | — |
| Cybench | 7.5% | — |

## Reasoning

- Llama 3.1-405B: 16.4 (#293)
- phi-3-medium 14B: —

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| BIG-Bench Hard | 82.9% | 81.4% |
| Epoch Capabilities Index | 128.76 | 121.18 |
| HellaSwag | 89.2% | 82.4% |
| WinoGrande | 89.2% | 81.5% |
| SimpleBench | 23% | — |
| Kagi LLM Benchmark | 45% | — |
| LMArena Hard Prompts | 1270 | — |
| DTBench | 61.4% | — |
| Adversarial NLI | — | 55.8% |
| ForecastBench | 59.9 | — |
| PIQA | 85.9% | — |

## Math

- Llama 3.1-405B: 17.3 (#283)
- phi-3-medium 14B: 25.2 (#249)

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| MATH Level 5 | 49.8% | 17.6% |
| OTIS Mock AIME 2024-2025 | 9.7% | — |
| Omni-MATH | 24.9% | — |
| LMArena Math | 1281 | — |

## Knowledge

- Llama 3.1-405B: 30.4 (#221)
- phi-3-medium 14B: 9.1 (#301)

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| GPQA Diamond | 50.9% | 27.6% |
| ARC (AI2) Challenge | 95.3% | 91.6% |
| MMLU | 84.5% | 78% |
| TriviaQA | 82.7% | 73.9% |
| MMLU-Pro | 72.3% | — |
| Confabulations | 17.6% | — |
| GPQA (HELM) | 52.2% | — |
| LMArena Expert | 1243 | — |
| OpenBookQA | — | 87.4% |

## Multilingual

- Llama 3.1-405B: 40.9 (#209)
- phi-3-medium 14B: —

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| LMArena Non-English | 1248 | — |
| LMArena Chinese | 1242 | — |
| LMArena French | 1278 | — |
| LMArena German | 1252 | — |
| LMArena Japanese | 1208 | — |
| LMArena Korean | 1184 | — |
| LMArena Russian | 1265 | — |
| LMArena Spanish | 1259 | — |

## Instruction Following

- Llama 3.1-405B: 66.1 (#209)
- phi-3-medium 14B: —

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| IFEval | 81.1% | — |
| LMArena Instruction Following | 1259 | — |

## Long Context

- Llama 3.1-405B: 38.4 (#192)
- phi-3-medium 14B: —

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| LMArena Longer Query | 1266 | — |

## Writing & Preference

- Llama 3.1-405B: 39.1 (#246)
- phi-3-medium 14B: —

| Benchmark | Llama 3.1-405B | phi-3-medium 14B |
|---|---|---|
| LMArena Text | 1284 | — |
| LMArena Creative Writing | 1262 | — |
| EQ-Bench Creative Writing | 870 | — |
| WildBench | 78.3% | — |
| LMArena Multi-Turn | 1297 | — |

## FAQ

### Is Llama 3.1-405B better than phi-3-medium 14B?

Llama 3.1-405B is the stronger model overall, scoring 30.4 to 29.1 on the Noometry Index.

### Is Llama 3.1-405B or phi-3-medium 14B better for coding?

phi-3-medium 14B scores higher on coding benchmarks: 36.3 versus 32.7 in the Noometry coding category.

### How many benchmarks do Llama 3.1-405B and phi-3-medium 14B share?

9 benchmarks have published results for both models. Llama 3.1-405B has 42 scored results on Noometry and phi-3-medium 14B has 13.