What are the hardware requirements for UGround V1 2B?

UGround V1 2B requires approximately 4.4 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed UGround V1 2B and how large is it?

UGround V1 2B is developed by osunlp, a model with 2b 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 UGround V1 2B?

UGround V1 2B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

UGround V1 2B by osunlp — VRAM 4.4GB, 32K context

Name: UGround V1 2B
Author: osunlp

UGround V1 2B is an open-source language model by osunlp. Features: 2b LLM, VRAM: 4.4GB, Context: 32K, License: apache-2.0.

Arxiv:2308.12966 Arxiv:2401.01614 Arxiv:2409.12191 Arxiv:2410.05243 Base model:finetune:qwen/qwen2... Base model:qwen/qwen2-vl-2b Conversational En Endpoints compatible Eval-results Image-text-to-text Multimodal Pytorch Qwen2 vl Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/osunlp/UGround-V1-2B

UGround V1 2B Parameters and Internals

LLM Name	UGround V1 2B
Repository 🤗	https://huggingface.co/osunlp/UGround-V1-2B
Base Model(s)	Qwen/Qwen2-VL-2B Qwen/Qwen2-VL-2B
Model Size	2b
Required VRAM	4.4 GB
Updated	2026-08-08
Maintainer	osunlp
Model Type	qwen2_vl
Model Files	4.4 GB 4.9 GB
Supported Languages	en
Model Architecture	Qwen2VLForConditionalGeneration
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.46.1
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Torch Data Type	bfloat16
Errors	replace

Rank the UGround V1 2B 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

Best Alternatives to UGround V1 2B

Best Alternatives	Context / RAM	Downloads	Likes
Qwen2 VL 2B Instruct	32K / 4.4 GB	3	24
ShowUI 2B	32K / 4.4 GB	3756	279
Qwen2 VL 2B Instruct Bf16	32K / 4.4 GB	90	1
Qwen2 VL 2B 4bit	32K / 1.2 GB	34	3
Qwen2 VL 2B Instruct 4bit	32K / 1.2 GB	3899	7
Qwen2 VL 2B 8bit	32K / 2.4 GB	21	1
Qwen2 VL 2B Instruct 8bit	32K / 2.4 GB	99	2
ShowUI 2B 6bit V2	32K / 2.8 GB	22	1
Qwen2 VL 2B Instruct AWQ	32K / 3 GB	3752	26
Qwen2 VL 2B Instruct GPTQ Int4	32K / 2.5 GB	793	28

Note: green Score (e.g. "73.2") means that the model is better than osunlp/UGround-V1-2B.