Model comparison
Phi-2 vs StarCoder 2 3B
Neither Phi-2 nor StarCoder 2 3B 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.
Side by side
| Phi-2 | StarCoder 2 3B | |
|---|---|---|
| Provider | Microsoft | NVIDIA |
| Noometry Index | — | 35.8 |
| Released | 2023-12-12 | 2024-02-22 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 11 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Phi-2: —, StarCoder 2 3B: 32.9
| Benchmark | Phi-2 | StarCoder 2 3B |
|---|---|---|
| HumanEval+ | 45.1% | 27.4% |
| BigCodeBench Complete | — | 21.4% |
| MBPP+ | 54.2% | — |
Reasoning Not comparable
Phi-2: —, StarCoder 2 3B: —
| Benchmark | Phi-2 | StarCoder 2 3B |
|---|---|---|
| Epoch Capabilities Index | 107.62 | 88.01 |
| WinoGrande | 54.7% | 57.1% |
| Adversarial NLI | 42.5% | — |
| BIG-Bench Hard | 59.4% | — |
| HellaSwag | 53.6% | — |
Math Not comparable
Phi-2: —, StarCoder 2 3B: —
| Benchmark | Phi-2 | StarCoder 2 3B |
|---|---|---|
| GSM8K | — | 21.6% |
Knowledge Not comparable
Phi-2: —, StarCoder 2 3B: —
| Benchmark | Phi-2 | StarCoder 2 3B |
|---|---|---|
| ARC (AI2) Challenge | 75.9% | 34.2% |
| MMLU | 58.4% | 36.6% |
| OpenBookQA | 73.6% | — |
| TriviaQA | 45.2% | — |
Frequently asked questions
Is Phi-2 better than StarCoder 2 3B?
Neither Phi-2 nor StarCoder 2 3B has enough public benchmark results to be ranked yet; the rows below show what has been published.
How many benchmarks do Phi-2 and StarCoder 2 3B share?
5 benchmarks have published results for both models. Phi-2 has 11 scored results on Noometry and StarCoder 2 3B has 7.