Getting Started¶
This guide helps you set up LM Studio and run your first local LLM script. Perfect for beginners exploring enterprise automation.
Setup Flow¶
flowchart TD
A[Download LM Studio] --> B[Install LM Studio]
B --> C[Download Qwen 2.5 Model]
C --> D[Start Local Server]
D --> E[Configure Port 1234]
E --> F[Run First Script]
style A fill:#e3f2fd
style D fill:#fff3e0
style F fill:#c8e6c9
Model Download Process¶
flowchart LR
A[LM Studio] --> B[Search Models]
B --> C[Download Qwen 2.5 1.5B]
C --> D[Load Model]
D --> E[Start Server]
style C fill:#fff3e0
style E fill:#c8e6c9
Script Execution Sequence¶
sequenceDiagram
participant S as Script
participant L as LM Studio
participant M as Model
S->>L: POST /v1/chat/completions
L->>M: Load Model
M-->>L: Model Ready
L-->>S: JSON Response
S->>S: Parse & Display
Prerequisites¶
- Windows 10/11 or macOS
- 8GB+ RAM recommended
- 4GB+ VRAM for 4-bit models
Quick Start Commands¶
# 1. Download LM Studio from lmstudio.ai
# 2. Search and download "Qwen 2.5 1.5B"
# 3. Start server (click "Start Server" button)
# 4. Run your first script
python llm_quick_demo_base.py
First Script Structure¶
graph TD
A[Import requests] --> B[Define API URL]
B --> C[Create Payload]
C --> D[Send POST Request]
D --> E[Parse Response]
E --> F[Display Output]
style A fill:#e8eaf6
style F fill:#c8e6c9
Next Steps¶
- Try ISP Classifier - Classify customer complaints
- Explore Qwen + RAG - Knowledge-augmented responses
- Check Gemma E4B - Advanced reasoning