Views
No views yet
| openPangu-VL-7B | |
|---|---|
| 语言模型 | |
| Architecture | Dense |
| Parameters (Non-Embedding) | 7B |
| Number of Layers | 34 |
| Hidden Dimension | 12800 |
| Attention Mechanism | GQA |
| Number of Attention Heads | 32 for Q，8 for KV |
| Vocabulary Size | 153k |
| Context Length (Natively) | 128k |
| 视觉编码器 | |
| Architecture | 22 Window Attention + 4 Full Attention |
| Number of Layers | 26 |
| Attention Hidden Size | 1536 |
| FFN Hidden Size | 4608 |
| Number of Attention Heads | 16 |
| Parameters | 615M |
| 测评集 | openPangu-VL-7B |
|---|---|
| 通用视觉对话 | |
| MMBench V1.1 DEV | 86.5 |
| MMStar | 70.1 |
| RealWorldQA | 76.1 |
| AI2D | 84.7 |
| 光学符号识别/文档理解 | |
| OCRBench | 907 |
| TextVQA | 85.1 |
| DocVQA | 96.0 |
| ChartQA | 88.3 |
| CharXiv dq/rq | 83.9/54.3 |
| 视觉学科能力 | |
| MMMU | 65.2 |
| MMMU-Pro | 52.6 |
| MathVista | 75.0 |
| 目标定位/计数 | |
| RefCOCO-avg | 90.6 |
| ODinW-13 | 51.5 |
| CountBench | 96.1 |
| Point-Bench | 65.4 |
| 多图 | |
| BLINK | 63.3 |
| MUIRBench | 61.6 |
| 视频理解 | |
| MVBench | 74.0 |
| VideoMME (w/o sub) | 68.0 |
| MLVU | 76.9 |
| 语言能力 | |
| MMLU-Pro | 78.2 |
| MMLU-Redux | 87.3 |
| GPQA-Diamond | 65.2 |
| C-Eval | 83.2 |
| AIME25 | 36.5 |
| Math-500 | 89.4 |
| LiveCodeBenchV6 | 24.6 |
| MBPP+ | 68.5 |
| IFEval | 83.0 |
cd inference/vllm_ascend/examples; python quick_start.pycd inference; pip install -r requirements.txtcd inference; python generate.pycookbooks。