Tutorials

This tutorial hub contains full-length guides (900–1400 words each) for safe Open-AutoGLM evaluation and deployment.

Why use a model service endpoint?

• Lower hardware barrier (no local GPU required)
• Faster setup (skip vLLM/SGLang install + large downloads)
• Easier updates and maintenance
• Recommended for beginners (aligns with official guidance)

Tutorials

1. Open-AutoGLM installation: prerequisites and first run
   A complete setup walkthrough, from Python and ADB to a first demo run.
2. ADB device not found: fix adb devices issues
   A troubleshooting playbook for USB, drivers, and permission prompts.
3. Run AutoGLM-Phone-9B-Multilingual for English apps
   How to choose the multilingual model and validate English UI flows.
4. Docker deployment checklist for Open-AutoGLM
   Containerization guide with safety and network defaults.
5. Human-in-the-loop design for logins and verification codes
   Patterns for approvals and safe checkpoints.
6. Phone Agent vs Appium: when LLM agents win or lose
   Neutral tradeoffs between agentic and deterministic automation.
7. Prevent destructive actions: confirmations and dry runs
   How to reduce risky actions without blocking workflows.
8. Benchmark primer: what “Phone Use” agents are measured on
   A grounded explanation of benchmark scope and limits.
9. Open-AutoGLM on low-VRAM hardware: options and tradeoffs
   Practical strategies for smaller GPUs or no-GPU setups.
10. Model service endpoints: connect and secure --base-url
    Security and reliability tips for external inference endpoints.