Model comparison
Phi-3.5-mini vs phi-3-medium 14B
phi-3-medium 14B has enough public results to be ranked (#306); Phi-3.5-mini does not yet, so treat this comparison as directional.
Last verified . 2 shared benchmarks.
MicrosoftSummary
- They share 2 benchmarks with published results for both. Phi-3.5-mini scores higher in 0 categories and phi-3-medium 14B in 1 category; one gap is clear of the uncertainty.
- The biggest single-benchmark swing is BigCodeBench Complete: 38.5% for Phi-3.5-mini and 48.7% for phi-3-medium 14B.
Side by side
| Phi-3.5-mini | phi-3-medium 14B | |
|---|---|---|
| Provider | Microsoft | Microsoft |
| Noometry Index | 36.9 | 29.7 |
| Released | 2024-08-16 | 2024-04-23 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 5 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding phi-3-medium 14B leads
Phi-3.5-mini: 34.6 (#232), phi-3-medium 14B: 36.8 (#201)
| Benchmark | Phi-3.5-mini | phi-3-medium 14B |
|---|---|---|
| BigCodeBench Instruct | 32.8% | 37.6% |
| BigCodeBench Complete | 38.5% | 48.7% |
Reasoning Not comparable
Phi-3.5-mini: —, phi-3-medium 14B: —
| Benchmark | Phi-3.5-mini | phi-3-medium 14B |
|---|---|---|
| Adversarial NLI | — | 55.8% |
| BIG-Bench Hard | — | 81.4% |
| Epoch Capabilities Index | — | 121.23 |
| HellaSwag | — | 82.4% |
| PIQA | 81% | — |
| WinoGrande | — | 81.5% |
Math Not comparable
Phi-3.5-mini: —, phi-3-medium 14B: 27.3 (#250)
| Benchmark | Phi-3.5-mini | phi-3-medium 14B |
|---|---|---|
| MATH Level 5 | — | 17.6% |
| GSM8K | 86.2% | — |
Knowledge Not comparable
Phi-3.5-mini: —, phi-3-medium 14B: 9.1 (#306)
| Benchmark | Phi-3.5-mini | phi-3-medium 14B |
|---|---|---|
| GPQA Diamond | — | 27.6% |
| ARC (AI2) Challenge | — | 91.6% |
| BoolQ | 78% | — |
| MMLU | — | 78% |
| OpenBookQA | — | 87.4% |
| TriviaQA | — | 73.9% |
Frequently asked questions
Is Phi-3.5-mini better than phi-3-medium 14B?
phi-3-medium 14B has enough public results to be ranked (#306); Phi-3.5-mini does not yet, so treat this comparison as directional.
Is Phi-3.5-mini or phi-3-medium 14B better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.8 versus 34.6 in the Noometry coding category.
How many benchmarks do Phi-3.5-mini and phi-3-medium 14B share?
2 benchmarks have published results for both models. Phi-3.5-mini has 5 scored results on Noometry and phi-3-medium 14B has 13.