LLM EXPLORER 63,791 MODELS INDEXED

MAI UI 2B 6bit V2 by mlx-community

By mlx-community · 74 downloads
HF Repo ↗ Copy URL

MAI UI 2B 6bit V2 is an open-source language model by mlx-community. Features: 2b LLM, VRAM: 2.2GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.22.

  Arxiv:2512.22047   2b   6-bit   6bit   Agent   Apple-silicon Base model:quantized:tongyi-ma... Base model:tongyi-mai/mai-ui-2...   Computer-use   Conversational   En   Grounding   Gui   Image-text-to-text   Mai-ui   Mlx   Mlx-vlm   Mobile-navigation   Quantized   Qwen3 vl   Region:us   Safetensors   Vision   Vision-language-model
Model Card on HF 🤗: https://huggingface.co/mlx-community/MAI-UI-2B-6bit-v2 

MAI UI 2B 6bit V2 Parameters and Internals

LLM NameMAI UI 2B 6bit V2
Repository 🤗https://huggingface.co/mlx-community/MAI-UI-2B-6bit-v2 
Base Model(s)  MAI UI 2B   Tongyi-MAI/MAI-UI-2B
Model Size2b
Required VRAM2.2 GB
Updated2026-10-11
Maintainermlx-community
Model Typeqwen3_vl
Model Files  2.2 GB
Supported Languagesen
Quantization Type6bit
Model ArchitectureQwen3VLForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version4.57.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to MAI UI 2B 6bit V2

Best Alternatives
Context / RAM
Downloads
Likes
...L 2B Instruct Unsloth Bnb 4bit0K / 2.4 GB1648448
...i Qwen3 VL 2B Instruct Ab 4bit0K / 1.8 GB1441
Qwen3 VL 2B Instruct 4bit0K / 1.8 GB18812
Qwen3 VL 2B Instruct Bnb 4bit0K / 2.2 GB9383
...L 2B Thinking Unsloth Bnb 4bit0K / 2.3 GB1771
Qwen3 VL 2B Instruct 8bit0K / 2.6 GB761
Qwen3 VL 2B Thinking 4bit0K / 1.8 GB401
Qwen3 VL 2B Instruct 6bit0K / 2.2 GB361
Qwen3 VL 2B Instruct0K / 4.3 GB2204577448
Qwen3 VL Reranker 2B0K / 4.3 GB764323212
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/MAI-UI-2B-6bit-v2.