Model comparison
Llama 2-34B vs Phi-1.5
Neither Llama 2-34B nor Phi-1.5 has enough public benchmark results to be ranked yet; the rows below show what has been published.
Last verified . 6 shared benchmarks.
Meta
MicrosoftSummary
- They share 6 benchmarks with published results for both.
- Phi-1.5 has downloadable open weights; the other is API-only.
Side by side
| Llama 2-34B | Phi-1.5 | |
|---|---|---|
| Provider | Meta | Microsoft |
| Noometry Index | — | — |
| Released | 2023-07-18 | 2023-09-11 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 10 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Reasoning Not comparable
Llama 2-34B: —, Phi-1.5: —
| Benchmark | Llama 2-34B | Phi-1.5 |
|---|---|---|
| Epoch Capabilities Index | 105.2 | 91.53 |
| WinoGrande | 76.7% | 73.4% |
| BIG-Bench Hard | 44.1% | — |
| HellaSwag | — | 47.6% |
| PIQA | 81.9% | — |
Math Not comparable
Llama 2-34B: —, Phi-1.5: —
| Benchmark | Llama 2-34B | Phi-1.5 |
|---|---|---|
| GSM8K | 42.2% | — |
Knowledge Not comparable
Llama 2-34B: —, Phi-1.5: —
| Benchmark | Llama 2-34B | Phi-1.5 |
|---|---|---|
| ARC (AI2) Challenge | 54.5% | 44.4% |
| BoolQ | 83.7% | 75.8% |
| MMLU | 62.6% | 37.6% |
| OpenBookQA | 58.2% | 37.2% |
| TriviaQA | 84.6% | — |
Frequently asked questions
Is Llama 2-34B better than Phi-1.5?
Neither Llama 2-34B nor Phi-1.5 has enough public benchmark results to be ranked yet; the rows below show what has been published.
How many benchmarks do Llama 2-34B and Phi-1.5 share?
6 benchmarks have published results for both models. Llama 2-34B has 10 scored results on Noometry and Phi-1.5 has 7.