Model comparison
Phi-2 vs phi-3-medium 14B
phi-3-medium 14B has enough public results to be ranked (#306); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 9 shared benchmarks.
MicrosoftSummary
- They share 9 benchmarks with published results for both.
Side by side
| Phi-2 | phi-3-medium 14B | |
|---|---|---|
| Provider | Microsoft | Microsoft |
| Noometry Index | — | 29.7 |
| Released | 2023-12-12 | 2024-04-23 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 11 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Phi-2: —, phi-3-medium 14B: 36.8 (#201)
| Benchmark | Phi-2 | phi-3-medium 14B |
|---|---|---|
| BigCodeBench Instruct | — | 37.6% |
| BigCodeBench Complete | — | 48.7% |
| HumanEval+ | 45.1% | — |
| MBPP+ | 54.2% | — |
Reasoning Not comparable
Phi-2: —, phi-3-medium 14B: —
| Benchmark | Phi-2 | phi-3-medium 14B |
|---|---|---|
| Adversarial NLI | 42.5% | 55.8% |
| BIG-Bench Hard | 59.4% | 81.4% |
| Epoch Capabilities Index | 107.94 | 121.23 |
| HellaSwag | 53.6% | 82.4% |
| WinoGrande | 54.7% | 81.5% |
Math Not comparable
Phi-2: —, phi-3-medium 14B: 27.3 (#250)
| Benchmark | Phi-2 | phi-3-medium 14B |
|---|---|---|
| MATH Level 5 | — | 17.6% |
Knowledge Not comparable
Phi-2: —, phi-3-medium 14B: 9.1 (#306)
| Benchmark | Phi-2 | phi-3-medium 14B |
|---|---|---|
| ARC (AI2) Challenge | 75.9% | 91.6% |
| MMLU | 58.4% | 78% |
| OpenBookQA | 73.6% | 87.4% |
| TriviaQA | 45.2% | 73.9% |
| GPQA Diamond | — | 27.6% |
Frequently asked questions
Is Phi-2 better than phi-3-medium 14B?
phi-3-medium 14B has enough public results to be ranked (#306); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Phi-2 and phi-3-medium 14B share?
9 benchmarks have published results for both models. Phi-2 has 11 scored results on Noometry and phi-3-medium 14B has 13.