Gpt2 vs Qwen2.5 0.5B Instruct
Lado a lado. Cada métrica. Cada benchmark.
| Tipo | Gpt2 | Qwen2.5 0.5B Instruct |
|---|---|---|
| Provider |  OpenAI |  Alibaba |
| puntuación promedio | 6.5 | 10.1 |
| Precio de entrada | - | - |
| Precio de salida | - | - |
| Ventana de contexto | - | - |
| Publicado el | 2022-03-02 | 2024-09-16 |
| Código abierto | Open Source | Open Source |
Puntuaciones de benchmark
6 benchmarks · Gpt2: 1, Qwen2.5 0.5B Instruct: 5
| Benchmark | Categoría | Gpt2 | Qwen2.5 0.5B Instruct |
|---|---|---|---|
| BBH (HuggingFace) | general | 2.7 | 8.2 |
| GPQA | knowledge | 1.1 | 1.2 |
| IFEval | language | 17.9 | 31.5 |
| MATH Level 5 | math | 0.2 | 10.3 |
| MMLU-PRO | knowledge | 1.8 | 8.0 |
| MUSR | reasoning | 15.3 | 1.4 |