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

AutoGLM-Phone-9B-Multilingual

👋 Join our WeChat community

⚠️ This project is intended for research and educational purposes only.
Any use for illegal data access, system interference, or unlawful activities is strictly prohibited.
Please review our Terms of Use carefully.

Project Overview

Phone Agent is a mobile intelligent assistant framework built on AutoGLM, capable of understanding smartphone screens through multimodal perception and executing automated operations to complete tasks.
The system controls devices via ADB (Android Debug Bridge), uses a vision-language model for screen understanding, and leverages intelligent planning to generate and execute action sequences.

Users can simply describe tasks in natural language—for example, “Open Xiaohongshu and search for food recommendations.”
Phone Agent will automatically parse the intent, understand the current UI, plan the next steps, and carry out the entire workflow.

The system also includes:

Sensitive action confirmation mechanisms
Human-in-the-loop fallback for login or verification code scenarios
Remote ADB debugging, allowing device connection via WiFi or network for flexible remote control and development

Model Usage

We provide an open-source model usage guide to help you quickly download and deploy the model.
Please visit our GitHub for detailed instructions.

The model architecture is identical to GLM-4.1V-9B-Thinking.
For deployment details, see the GLM-V repository.

Citation

If you find our work helpful, please cite the following paper:

bibtex

@article{liu2024autoglm,
  title={Autoglm: Autonomous foundation agents for guis},
  author={Liu, Xiao and Qin, Bo and Liang, Dongzhu and Dong, Guang and Lai, Hanyu and Zhang, Hanchen and Zhao, Hanlin and Iong, Iat Long and Sun, Jiadai and Wang, Jiaqi and others},
  journal={arXiv preprint arXiv:2411.00820},
  year={2024}
}
@article{xu2025mobilerl,
  title={MobileRL: Online Agentic Reinforcement Learning for Mobile GUI Agents},
  author={Xu, Yifan and Liu, Xiao and Liu, Xinghan and Fu, Jiaqi and Zhang, Hanchen and Jing, Bohao and Zhang, Shudan and Wang, Yuting and Zhao, Wenyi and Dong, Yuxiao},
  journal={arXiv preprint arXiv:2509.18119},
  year={2025}
}

🚀 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/AutoGLM-Phone-9B-Multilingual-GGUF

AutoGLM-Phone-9B-Multilingual GGUF Models

Model Generation Details

Quantization Beyond the IMatrix