# Llama 3-70B vs Phi-1.5

> Llama 3-70B has enough public results to be ranked (#323); Phi-1.5 does not yet, so treat this comparison as directional.

- Canonical page: https://noometry.com/compare/llama-3-70b-vs-phi-1-5
- Last updated: 2026-10-11
- Shared benchmarks: 3

## Summary

- They share 3 benchmarks with published results for both.

## Snapshot

| | Llama 3-70B | Phi-1.5 |
|---|---|---|
| Provider | Meta | Microsoft |
| Noometry Index | 28.8 | — |
| Rank | 323 | — |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Llama 3-70B: 35.8 (#218)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| BigCodeBench Instruct | 43.6% | — |
| LMArena Coding | 1206 | — |
| BigCodeBench Complete | 54.5% | — |
| HumanEval+ | 72% | — |
| MBPP+ | 69% | — |

## Agentic & Tool Use

- Llama 3-70B: 21.1 (#139)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| Cybench | 5% | — |

## Reasoning

- Llama 3-70B: 18.0 (#288)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| Epoch Capabilities Index | 122.93 | 91.53 |
| WinoGrande | 83.5% | 73.4% |
| Kagi LLM Benchmark | 35.1% | — |
| LMArena Hard Prompts | 1195 | — |
| DTBench | 54.2% | — |
| ForecastBench | 57.1 | — |
| HellaSwag | — | 47.6% |

## Math

- Llama 3-70B: 12.8 (#305)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | — |
| LMArena Math | 1218 | — |
| MATH Level 5 | 22.6% | — |

## Knowledge

- Llama 3-70B: 20.8 (#277)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| MMLU | 79.3% | 37.6% |
| GPQA Diamond | 40.6% | — |
| LMArena Expert | 1149 | — |
| ARC (AI2) Challenge | — | 44.4% |
| BoolQ | — | 75.8% |
| OpenBookQA | — | 37.2% |

## Multilingual

- Llama 3-70B: 33.6 (#251)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| LMArena Non-English | 1142 | — |
| LMArena Chinese | 1114 | — |
| LMArena French | 1232 | — |
| LMArena German | 1169 | — |
| LMArena Japanese | 1017 | — |
| LMArena Korean | 1017 | — |
| LMArena Russian | 1159 | — |
| LMArena Spanish | 1241 | — |

## Instruction Following

- Llama 3-70B: 62.5 (#238)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| LMArena Instruction Following | 1194 | — |

## Long Context

- Llama 3-70B: 35.6 (#240)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| LMArena Longer Query | 1174 | — |

## Writing & Preference

- Llama 3-70B: 42.8 (#231)
- Phi-1.5: —

| Benchmark | Llama 3-70B | Phi-1.5 |
|---|---|---|
| LMArena Text | 1221 | — |
| LMArena Creative Writing | 1210 | — |
| LMArena Multi-Turn | 1223 | — |

## FAQ

### Is Llama 3-70B better than Phi-1.5?

Llama 3-70B has enough public results to be ranked (#323); Phi-1.5 does not yet, so treat this comparison as directional.

### How many benchmarks do Llama 3-70B and Phi-1.5 share?

3 benchmarks have published results for both models. Llama 3-70B has 31 scored results on Noometry and Phi-1.5 has 7.