microsoft/vidur

A large-scale simulation framework for LLM inference

52
/ 100
Established

This project helps operations engineers and AI infrastructure managers understand how different large language models will perform on various hardware setups and under varying user demand. You input model details, hardware configurations, and anticipated user request patterns, and it outputs performance metrics like response times and throughput. This allows you to plan capacity and optimize your LLM deployments.

547 stars. No commits in the last 6 months.

Use this if you need to determine the best hardware and software configurations for deploying Large Language Models without costly physical testing.

Not ideal if you are looking for a tool to train LLMs or to benchmark the accuracy of different models.

LLM-deployment capacity-planning performance-engineering AI-infrastructure system-simulation
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 24 / 25

How are scores calculated?

Stars

547

Forks

104

Language

Python

License

MIT

Last pushed

Jul 25, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/microsoft/vidur"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.