---
title: LLM Inference Simulator — Extended
emoji: 🚀
colorFrom: gray
colorTo: purple
sdk: static
pinned: false
---

# LLM Inference — Pipeline Simulator v2

Extended version covering concurrency, batching, prefix caching, and memory constraints. From EXD Episode 4 — Performance Tuning.

Built with vanilla HTML/CSS/JS.

📺 [Watch the episode](https://youtube.com/@EXD-ai) | 💻 [GitHub](https://github.com/Ramshreyas/EXD)