Fine-tune your own model (LoRA / QLoRA) | Sri AI

Course overview

Make an open model yours: teach it your subject, your tone and your language on a single GPU.

Level: Advanced  ·  Mode: Part-time

Who this course is for

Developers who need a model that knows their domain or speaks their language properly.

What you will learn

• Decide between fine-tuning and retrieval
• Build a clean instruction dataset
• Fine-tune with LoRA and QLoRA
• Export and run your model in Ollama

Syllabus

1. Module 1: Fine-tuning versus retrieval
2. Module 2: Building an instruction dataset
3. Module 3: LoRA and QLoRA
4. Module 4: Training with Unsloth
5. Module 5: Sinhala and Tamil adaptation
6. Module 6: Evaluating before and after
7. Module 7: Exporting to GGUF

Final project

Every module ends in hands-on practice, and the course ends with a project you build and present. Your certificate names that project.

Before you start

Run your own AI and Python for AI.

Open-source tools you will use

• Unsloth
• TRL
• Smol Course

Adapts material from the Hugging Face Smol Course (Apache-2.0).