Key Specifications
| Vendor | alibaba |
|---|
| Version | 2-1-5b |
|---|
| Release Date | 2024-06-07 |
|---|
| Context Window | 32768 tokens |
|---|
| Input Modalities | text |
|---|
| Output Modalities | text |
|---|
| License | Qwen License |
|---|
| Documentation | https://qwenlm.github.io/ |
|---|
Benchmark Performance
| Benchmark | Score | Unit | Evaluated At | Notes | Source |
|---|
| MMLU | 40.5 | % | 2024-06-07 | 5-shot | view |
| HUMANEVAL | 47.2 | pass@1 | 2024-06-07 | — | view |
| GSM8K | 36.2 | % | 2024-06-07 | 0-shot CoT | view |
| MATH | 25.4 | % | 2024-06-07 | 0-shot CoT | view |
| BBH | 51.5 | % | 2024-06-07 | 3-shot CoT | view |
| GPQA | 22.1 | % | 2024-06-07 | 0-shot | view |
| IFEVAL | 57.4 | % | 2024-06-07 | prompt_strict | view |
| ARC | 85.4 | % | 2024-06-07 | challenge | view |
| MUSR | 27.8 | % | 2024-06-07 | 0-shot | view |
| WINOGRANDE | 62.9 | % | 2024-06-07 | 0-shot | view |
Compliance
- Data Residency: CN
- SOC2: ✗
- HIPAA: ✗
- GDPR: ✗
- ISO 27001: ✗
Qwen2 1.5B
Panoramica del modello
阿里巴巴 Qwen2 1.5B 轻量开源模型, 32K 上下文, 1.5B 参数, 适合边缘设备与移动部署。
Specifiche principali
| Fornitore | Versione | Data di rilascio | Finestra di contesto | Modalità di input | Modalità di output | Licenza |
|---|
| Alibaba | 2-1-5b | 2024-06-07 | 32K | text | text | Qwen License |
Prestazioni benchmark
| Benchmark | Punteggio | Unità | Note |
|---|
| MMLU (Massive Multitask Language Understanding) | 40.5 | % | 5-shot |
| HumanEval | 47.2 | pass@1 | — |
| GSM8K (Grade School Math 8K) | 36.2 | % | 0-shot CoT |
| MATH | 25.4 | % | 0-shot CoT |
| BBH (BIG-Bench Hard) | 51.5 | % | 3-shot CoT |
| GPQA | 22.1 | % | 0-shot |
| IFEval | 57.4 | % | prompt_strict |
| ARC | 85.4 | % | challenge |
| MUSR | 27.8 | % | 0-shot |
| WinoGrande | 62.9 | % | 0-shot |
Prezzi
| Input | Output | Lettura cache | Scrittura cache |
|---|
| — | — | — | — |
per milione di token
Punti di forza
Punti deboli
- MMLU 仅 40.5，知识推理偏弱。
- HumanEval 47.2，代码能力较弱。
- 闭源专有模型，不支持自托管。
Casi d’uso
Riferimenti