Skip to main content

Module 3: Running an LLM on a GPU

— A collection of 14 posts
September 2026

Putting It All Together

September 2026

Multi-GPU Inference

September 2026

Beyond Batching

September 2026

PagedAttention

September 2026

Continuous Batching

September 2026

Batching

September 2026

Prefill vs Decode

September 2026

Decode

September 2026

Reducing the KV Cache

September 2026

The KV Cache

September 2026

Attention: How Tokens Use Other Tokens

September 2026

What Actually Moves Through the Model?

September 2026

Loading a Model & the VRAM Budget

September 2026

The Whole Path