Model comparison
Llama 2-70B vs Phi-2
Llama 2-70B has enough public results to be ranked (#349); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 8 shared benchmarks.
Meta
MicrosoftSummary
- They share 8 benchmarks with published results for both.
Side by side
| Llama 2-70B | Phi-2 | |
|---|---|---|
| Provider | Meta | Microsoft |
| Noometry Index | 24.4 | — |
| Released | 2023-07-18 | 2023-12-12 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Llama 2-70B: 31.4 (#286), Phi-2: —
| Benchmark | Llama 2-70B | Phi-2 |
|---|---|---|
| LMArena Coding | 1079 | — |
| HumanEval+ | — | 45.1% |
| MBPP+ | — | 54.2% |
Reasoning Not comparable
Llama 2-70B: 14.4 (#325), Phi-2: —
| Benchmark | Llama 2-70B | Phi-2 |
|---|---|---|
| BIG-Bench Hard | 64.9% | 59.4% |
| Epoch Capabilities Index | 113.79 | 107.94 |
| HellaSwag | 85.3% | 53.6% |
| WinoGrande | 80.2% | 54.7% |
| LMArena Hard Prompts | 1073 | — |
| DTBench | 41.6% | — |
| Adversarial NLI | — | 42.5% |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
Math Not comparable
Llama 2-70B: 8.1 (#326), Phi-2: —
| Benchmark | Llama 2-70B | Phi-2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0% | — |
| LMArena Math | 1091 | — |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |
Knowledge Not comparable
Llama 2-70B: 7.4 (#310), Phi-2: —
| Benchmark | Llama 2-70B | Phi-2 |
|---|---|---|
| ARC (AI2) Challenge | 78.3% | 75.9% |
| MMLU | 69.9% | 58.4% |
| OpenBookQA | 60.2% | 73.6% |
| TriviaQA | 87.6% | 45.2% |
| GPQA Diamond | 26.3% | — |
| LMArena Expert | 1039 | — |
| BoolQ | 88.6% | — |
Multilingual Not comparable
Llama 2-70B: 27.7 (#274), Phi-2: —
| Benchmark | Llama 2-70B | Phi-2 |
|---|---|---|
| LMArena Non-English | 1045 | — |
| LMArena Chinese | 995 | — |
| LMArena French | 1090 | — |
| LMArena German | 1041 | — |
| LMArena Japanese | 927 | — |
| LMArena Korean | 964 | — |
| LMArena Russian | 1083 | — |
| LMArena Spanish | 1143 | — |
Instruction Following Not comparable
Llama 2-70B: 54.9 (#278), Phi-2: —
| Benchmark | Llama 2-70B | Phi-2 |
|---|---|---|
| LMArena Instruction Following | 1071 | — |
Long Context Not comparable
Llama 2-70B: 32.3 (#270), Phi-2: —
| Benchmark | Llama 2-70B | Phi-2 |
|---|---|---|
| LMArena Longer Query | 1062 | — |
Writing & Preference Not comparable
Llama 2-70B: 32.3 (#279), Phi-2: —
| Benchmark | Llama 2-70B | Phi-2 |
|---|---|---|
| LMArena Text | 1115 | — |
| LMArena Creative Writing | 1075 | — |
| LMArena Multi-Turn | 1088 | — |
Frequently asked questions
Is Llama 2-70B better than Phi-2?
Llama 2-70B has enough public results to be ranked (#349); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Llama 2-70B and Phi-2 share?
8 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Phi-2 has 11.