I've been experimenting with a new quantization approach that selectively elevates the precision of key layers beyond what the default IMatrix configuration provides.

In my testing, standard IMatrix quantization underperforms at lower bit depths, especially with Mixture of Experts (MoE) models. To address this, I'm using the --tensor-type option in llama.cpp to manually "bump" important layers to higher precision. You can see the implementation here:
👉 Layer bumping with llama.cpp

While this does increase model file size, it significantly improves precision for a given quantization level.

I'd love your feedback—have you tried this? How does it perform for you?

Click here to get info on choosing the right GGUF model format

BU-30B-A3B-Preview

Meet BU-30B-A3B-Preview — bringing SoTA Browser Use capabilities in a small model that can be hosted on a single GPU.

This model is heavily trained to be used with browser-use OSS library and provides comprehensive browsing capabilities with superior DOM understanding and visual reasoning.

Quickstart (BU Cloud)

You can directly use this model at BU Cloud. Simply

Get your API key from BU Cloud
Set environment variable: export BROWSER_USE_API_KEY="your-key"
Install the browser-use library following the instructions here and run

python

from dotenv import load_dotenv
from browser_use import Agent, ChatBrowserUse
load_dotenv()

llm = ChatBrowserUse(
	model='browser-use/bu-30b-a3b-preview',  # BU Open Source Model!!
)

agent = Agent(
	task='Find the number of stars of browser-use and stagehand. Tell me which one has more stars :)',
	llm=llm,
    flash_mode=True
)
agent.run_sync()

Quickstart (vLLM)

We recommend using this model with vLLM.

Installation

Make sure to install vllm >= 0.12.0:

pip install vllm --upgrade

Serve

A simple launch command is:

bash

vllm serve browser-use/bu-30b-a3b-preview \
  --max-model-len 65536 \
  --host 0.0.0.0 \
  --port 8000

which will create an OpenAI compatible endpoint at localhost that you can use with.

python

from dotenv import load_dotenv
from browser_use import Agent, ChatOpenAI
load_dotenv()

llm = ChatOpenAI(
    base_url='http://localhost:8000/v1',
    model='browser-use/bu-30b-a3b-preview',
    temperature=0.6,
    top_p=0.95,
    dont_force_structured_output=True, # speed up by disabling structured output
)

agent = Agent(
    task='Find the number of stars of browser-use and stagehand. Tell me which one has more stars :)',
    llm=llm,
)
agent.run_sync()

Model Details

Property	Value
Base Model	Qwen/Qwen3-VL-30B-A3B-Instruct
Parameters	30B total, 3B active (MoE)
Context Length	65,536 tokens
Architecture	Vision-Language Model (Mixture of Experts)

🚀 If you find these models useful

Help me test my AI-Powered Quantum Network Monitor Assistant with quantum-ready security checks:

👉 Quantum Network Monitor

The full Open Source Code for the Quantum Network Monitor Service available at my github repos ( repos with NetworkMonitor in the name) : Source Code Quantum Network Monitor. You will also find the code I use to quantize the models if you want to do it yourself GGUFModelBuilder

💬 How to test:
Choose an AI assistant type:

TurboLLM (GPT-4.1-mini)
HugLLM (Hugginface Open-source models)
TestLLM (Experimental CPU-only)

What I’m Testing

I’m pushing the limits of small open-source models for AI network monitoring, specifically:

Function calling against live network services
How small can a model go while still handling:
- Automated Nmap security scans
- Quantum-readiness checks
- Network Monitoring tasks

🟡 TestLLM – Current experimental model (llama.cpp on 2 CPU threads on huggingface docker space):

✅ Zero-configuration setup
⏳ 30s load time (slow inference but no API costs) . No token limited as the cost is low.
🔧 Help wanted! If you’re into edge-device AI, let’s collaborate!

Other Assistants

🟢 TurboLLM – Uses gpt-4.1-mini :

**It performs very well but unfortunatly OpenAI charges per token. For this reason tokens usage is limited.
Create custom cmd processors to run .net code on Quantum Network Monitor Agents
Real-time network diagnostics and monitoring
Security Audits
Penetration testing (Nmap/Metasploit)

🔵 HugLLM – Latest Open-source models:

🌐 Runs on Hugging Face Inference API. Performs pretty well using the lastest models hosted on Novita.

💡 Example commands you could test:

"Give me info on my websites SSL certificate"
"Check if my server is using quantum safe encyption for communication"
"Run a comprehensive security audit on my server"
'"Create a cmd processor to .. (what ever you want)" Note you need to install a Quantum Network Monitor Agent to run the .net code on. This is a very flexible and powerful feature. Use with caution!

Final Word

I fund the servers used to create these model files, run the Quantum Network Monitor service, and pay for inference from Novita and OpenAI—all out of my own pocket. All the code behind the model creation and the Quantum Network Monitor project is open source. Feel free to use whatever you find helpful.

If you appreciate the work, please consider buying me a coffee ☕. Your support helps cover service costs and allows me to raise token limits for everyone.

I'm also open to job opportunities or sponsorship.

Thank you! 😊

Mungert/bu-30b-a3b-preview-GGUF

bu-30b-a3b-preview GGUF Models

Model Generation Details

Quantization Beyond the IMatrix