Deepseek Vl2 8bit is an open-source language model by mlx-community. Features: 7.8b LLM, VRAM: 29.3GB, License: other, Quantized, LLM Explorer Score: 0.16.
| LLM Name | Deepseek Vl2 8bit |
| Repository π€ | https://huggingface.co/mlx-community/deepseek-vl2-8bit |
| Model Size | 7.8b |
| Required VRAM | 29.3 GB |
| Updated | 2026-08-08 |
| Maintainer | mlx-community |
| Model Type | deepseek_vl_v2 |
| Model Files | |
| Quantization Type | 8bit |
| Model Architecture | AutoModel |
| License | other |
| Transformers Version | 4.38.2 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <ο½βpadβο½> |
| Torch Data Type | bfloat16 |