Model comparison
Nemotron-4 15B vs Phi-1.5
Neither Nemotron-4 15B nor Phi-1.5 has enough public benchmark results to be ranked yet; the rows below show what has been published.
Last verified . 5 shared benchmarks.
Summary
- They share 5 benchmarks with published results for both.
- Phi-1.5 has downloadable open weights; the other is API-only.
Side by side
| Nemotron-4 15B | Phi-1.5 | |
|---|---|---|
| Provider | NVIDIA | Microsoft |
| Noometry Index | — | — |
| Released | 2024-02-26 | 2023-09-11 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 8 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Reasoning Not comparable
Nemotron-4 15B: —, Phi-1.5: —
| Benchmark | Nemotron-4 15B | Phi-1.5 |
|---|---|---|
| Epoch Capabilities Index | 107.38 | 90.82 |
| HellaSwag | 82.4% | 47.6% |
| WinoGrande | 78% | 73.4% |
| BIG-Bench Hard | 58.7% | — |
| PIQA | 82.4% | — |
Math Not comparable
Nemotron-4 15B: —, Phi-1.5: —
| Benchmark | Nemotron-4 15B | Phi-1.5 |
|---|---|---|
| GSM8K | 46% | — |
Knowledge Not comparable
Nemotron-4 15B: —, Phi-1.5: —
| Benchmark | Nemotron-4 15B | Phi-1.5 |
|---|---|---|
| ARC (AI2) Challenge | 55.5% | 44.4% |
| MMLU | 58.7% | 37.6% |
| BoolQ | — | 75.8% |
| OpenBookQA | — | 37.2% |
Frequently asked questions
Is Nemotron-4 15B better than Phi-1.5?
Neither Nemotron-4 15B 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 Nemotron-4 15B and Phi-1.5 share?
5 benchmarks have published results for both models. Nemotron-4 15B has 8 scored results on Noometry and Phi-1.5 has 7.