What are the hardware requirements for Bu 30B A3b Preview?

Bu 30B A3b Preview requires approximately 62.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Bu 30B A3b Preview and how large is it?

Bu 30B A3b Preview is developed by browser-use, a model with 30b 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 Bu 30B A3b Preview?

Bu 30B A3b Preview is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Bu 30B A3b Preview by browser-use — VRAM 62.3GB

Name: Bu 30B A3b Preview
Author: browser-use

Bu 30B A3b Preview is an open-source language model by browser-use. Features: 30b LLM, VRAM: 62.3GB, Instruction-Based, LLM Explorer Score: 0.27.

Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-vl-30b-a... Browser use Conversational En Endpoints compatible Image-text-to-text Instruct Qwen3 vl moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/browser-use/bu-30b-a3b-preview

Bu 30B A3b Preview Parameters and Internals

LLM Name	Bu 30B A3b Preview
Repository 🤗	https://huggingface.co/browser-use/bu-30b-a3b-preview
Base Model(s)	Qwen3 VL 30B A3B Instruct Qwen/Qwen3-VL-30B-A3B-Instruct
Model Size	30b
Required VRAM	62.3 GB
Updated	2026-08-08
Maintainer	browser-use
Model Type	qwen3_vl_moe
Instruction-Based	Yes
Model Files	4.2 GB: 1-of-13 5.0 GB: 2-of-13 5.0 GB: 3-of-13 5.0 GB: 4-of-13 5.0 GB: 5-of-13 5.0 GB: 6-of-13 5.0 GB: 7-of-13 5.0 GB: 8-of-13 5.0 GB: 9-of-13 5.0 GB: 10-of-13 5.0 GB: 11-of-13 5.0 GB: 12-of-13 3.1 GB: 13-of-13
Supported Languages	en
Model Architecture	Qwen3VLMoeForConditionalGeneration
Model Max Length	262144
Transformers Version	4.57.3
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Errors	replace

Rank the Bu 30B A3b Preview 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

Quantized Models of the Bu 30B A3b Preview

Model	Likes	Downloads	VRAM
Bu 30B A3b Preview 4bit	4	52	18 GB

Best Alternatives to Bu 30B A3b Preview

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3 VL 30B A3B Instruct	0K / 62.3 GB	488106	591
Qwen3 VL 30B A3B Instruct FP8	0K / 32.2 GB	734631	115
Qwen3 VL 30B A3B Instruct	0K / 62.3 GB	515	3
Gelato 30B A3B	0K / 62.1 GB	30	34
Qwen3 VL 30B A3B Instruct Bf16	0K / 62.3 GB	87	4
Qwen3 VL 30B A3B Instruct 4bit	0K / 18.3 GB	925	8
Qwen3 VL 30B A3B Instruct 8bit	0K / 33.5 GB	231	3
Qwen3 VL 30B A3B Instruct 6bit	0K / 25.8 GB	72	2
Qwen3 VL 30B A3B Instruct AWQ	0K / 17.9 GB	1184704	45

Note: green Score (e.g. "73.2") means that the model is better than browser-use/bu-30b-a3b-preview.