LLM EXPLORER 56,604 MODELS INDEXED

Deepseek Vl2 8bit by mlx-community

By mlx-community · 143 downloads

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.

  8bit   Conversational   Deepseek vl v2   Endpoints compatible   Image-text-to-text   Mlx   Quantized   Region:us   Safetensors   Sharded   Tensorflow

Deepseek Vl2 8bit Parameters and Internals

LLM NameDeepseek Vl2 8bit
Repository πŸ€—https://huggingface.co/mlx-community/deepseek-vl2-8bit 
Model Size7.8b
Required VRAM29.3 GB
Updated2026-08-08
Maintainermlx-community
Model Typedeepseek_vl_v2
Model Files  5.2 GB: 1-of-6   5.1 GB: 2-of-6   5.2 GB: 3-of-6   5.1 GB: 4-of-6   5.1 GB: 5-of-6   3.6 GB: 6-of-6
Quantization Type8bit
Model ArchitectureAutoModel
Licenseother
Transformers Version4.38.2
Tokenizer ClassLlamaTokenizer
Padding Token<ο½œβ–padβ–ο½œ>
Torch Data Typebfloat16