What are the hardware requirements for Deepseek Vl2 4bit?

Deepseek Vl2 4bit requires approximately 15.6 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Deepseek Vl2 4bit and how large is it?

Deepseek Vl2 4bit is developed by mlx-community, a model with 4.4b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Deepseek Vl2 4bit?

Deepseek Vl2 4bit is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Deepseek Vl2 4bit by mlx-community — VRAM 15.6GB

Name: Deepseek Vl2 4bit
Author: mlx-community

Deepseek Vl2 4bit is an open-source language model by mlx-community. Features: 4.4b LLM, VRAM: 15.6GB, License: other, Quantized.

4bit Conversational Deepseek vl v2 Endpoints compatible Image-text-to-text Mlx Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mlx-community/deepseek-vl2-4bit

Deepseek Vl2 4bit Parameters and Internals

LLM Name	Deepseek Vl2 4bit
Repository 🤗	https://huggingface.co/mlx-community/deepseek-vl2-4bit
Model Size	4.4b
Required VRAM	15.6 GB
Updated	2026-08-08
Maintainer	mlx-community
Model Type	deepseek_vl_v2
Model Files	5.3 GB: 1-of-3 5.2 GB: 2-of-3 5.1 GB: 3-of-3
Quantization Type	4bit
Model Architecture	AutoModel
License	other
Transformers Version	4.38.2
Tokenizer Class	LlamaTokenizer
Padding Token	<｜▁pad▁｜>
Torch Data Type	bfloat16

Rank the Deepseek Vl2 4bit capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation