Model comparison
Phi-2 vs Phi-4 Mini
Phi-4 Mini has enough public results to be ranked (#283); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
MicrosoftSide by side
| Phi-2 | Phi-4 Mini | |
|---|---|---|
| Provider | Microsoft | Microsoft |
| Noometry Index | — | 30.9 |
| Released | 2023-12-12 | 2024-12-11 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.075 |
| Output $ / M tokens | — | $0.30 |
| Results tracked | 11 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Phi-2: —, Phi-4 Mini: 28.1 (#317)
| Benchmark | Phi-2 | Phi-4 Mini |
|---|---|---|
| SciCode | — | 10.8% |
| HumanEval+ | 45.1% | — |
| MBPP+ | 54.2% | — |
Reasoning Not comparable
Phi-2: —, Phi-4 Mini: 22.4 (#195)
| Benchmark | Phi-2 | Phi-4 Mini |
|---|---|---|
| CritPt | — | 0% |
| Adversarial NLI | 42.5% | — |
| BIG-Bench Hard | 59.4% | — |
| Epoch Capabilities Index | 107.94 | — |
| HellaSwag | 53.6% | — |
| WinoGrande | 54.7% | — |
Knowledge Not comparable
Phi-2: —, Phi-4 Mini: 25.3 (#262)
| Benchmark | Phi-2 | Phi-4 Mini |
|---|---|---|
| Vectara Hallucination Rate | — | 23.5% |
| ARC (AI2) Challenge | 75.9% | — |
| MMLU | 58.4% | — |
| OpenBookQA | 73.6% | — |
| TriviaQA | 45.2% | — |
Frequently asked questions
Is Phi-2 better than Phi-4 Mini?
Phi-4 Mini has enough public results to be ranked (#283); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Phi-2 and Phi-4 Mini share?
0 benchmarks have published results for both models. Phi-2 has 11 scored results on Noometry and Phi-4 Mini has 3.